Hydrogen bonding and catalysis of solvolysis of 4-methoxybenzyl fluoride

Article abstract of DOI:10.1021/ja026849s

Values of k_o = 8.0 × 10^-3 s^-1 and k_H = 2.5 × 10^-2 M^-1 s^-1, respectively, were determined for the spontaneous and the acid-catalyzed cleavage of 4-methoxybenzyl fluoride (1-F) to form the 4-methoxybenzyl carbocation (1⁺). Values of k_F = 1.8 × 10⁷ M^-1 s^-1 and k_HF = 7.2 × 10⁴ M^-1 s^-1 were determined for addition of F^- and HF to 1⁺ for reaction in the microscopic reverse direction. Evidence is presented that the reversible addition of HF to 1⁺ to give 1-F + H⁺ proceeds by a concerted reaction mechanism. The relatively small 250-fold difference between the reactivities of fluoride ion and neutral HF toward 1⁺ is attributed to the tendency of the strong aqueous solvation of F^- to decrease its nucleophilic reactivity and to the advantage for the concerted compared with the usual stepwise pathway for addition of HF. There is no significant stabilization of the transition state for cleavage of 1-F from general acid catalysis by 0.80 M cyanoacetate buffer at pH 1.7. The estimated 3 kcal/mol larger Marcus intrinsic barrier for heterolytic cleavage of 1-F than for cleavage of 1-Cl is attributed to a lag in the development at the transition state of the ca. 30 kcal/mol greater stabilizing solvation of the product ion F^- compared with Cl^-. The decrease in the electronegativity of X along the series X = F, OH, Cl is accompanied by a ca. 101⁰-fold increase in the carbon basicity compared with the proton basicity of X^-.

Full text of DOI:10.1021/ja026849s

Published on Web 07/25/2002
Hydrogen Bonding and Catalysis of Solvolysis of
4-Methoxybenzyl Fluoride
Maria M. Toteva and John P. Richard*
Contribution from the Department of Chemistry, UniVersity at Buffalo, SUNY,
Buffalo, New York 14260-3000
Received May 9, 2002
Abstract: Values of ko ) 8.0 × 10-3 s-1 and kH ) 2.5 × 10-2 M-1 s-1, respectively, were determined for
the spontaneous and the acid-catalyzed cleavage of 4-methoxybenzyl fluoride (1-F) to form the
4-methoxybenzyl carbocation (1+). Values of kF ) 1.8 × 107 M-1 s-1 and kHF ) 7.2 × 104 M-1 s-1 were
determined for addition of F- and HF to 1+ for reaction in the microscopic reverse direction. Evidence is
presented that the reversible addition of HF to 1+ to give 1-F + H+ proceeds by a concerted reaction
mechanism. The relatively small 250-fold difference between the reactivities of fluoride ion and neutral HF
toward 1+ is attributed to the tendency of the strong aqueous solvation of F- to decrease its nucleophilic
reactivity and to the advantage for the concerted compared with the usual stepwise pathway for addition
of HF. There is no significant stabilization of the transition state for cleavage of 1-F from general acid
catalysis by 0.80 M cyanoacetate buffer at pH 1.7. The estimated 3 kcal/mol larger Marcus intrinsic barrier
for heterolytic cleavage of 1-F than for cleavage of 1-Cl is attributed to a lag in the development at the
transition state of the ca. 30 kcal/mol greater stabilizing solvation of the product ion F- compared with Cl-.
The decrease in the electronegativity of X along the series X ) F, OH, Cl is accompanied by a ca. 1010-
fold increase in the carbon basicity compared with the proton basicity of X-.
Scheme 1
Introduction
There is much interest in the effects of hydrogen bonding on
chemical reactivity1,2 and speculation about the contribution of
hydrogen bonding to the catalytic rate acceleration for enzyme
catalysts.3-5 These effects will be largest in reactions where
strong hydrogen bonds are formed or cleaved, such as heterolytic
C-F bond cleavage to form fluoride ion, a strong hydrogen
bond acceptor.2 The strong solvation of product fluoride ion6
will result in an increase in the overall equilibrium constant for
C-F bond cleavage, and this effect will be expressed in the
rate constants for cleavage and synthesis of alkyl fluorides. Also,
the transition state for the cleavage reaction can be stabilized
by proton transfer from hydronium ion (specific acid catalysis)
and possibly from general acids (general acid catalysis).
Specific acid catalysis has been reported for the solvolysis
in aqueous solution of simple alkyl fluorides,7,8 benzyl fluoride,9
triphenylmethyl fluoride,10,11 and R-D-glucosyl fluoride.12 How-
ever, the question of the effect of the strong solvation of fluoride
ion on the chemical reactivity of alkyl fluorides has not been
specifically addressed. A comparison of the reactivity toward
solvolysis of alkyl fluorides and related compounds with leaving
groups that form substantially weaker hydrogen bonds to solvent
than does fluoride ion should provide interesting insight into
the effects of hydrogen bonding on the rate of C-X bond
cleavage. 4-Methoxybenzyl derivatives (1-Nu, Scheme 1) are
an attractive candidate for this study because these compounds
are sufficiently soluble and reactive for a room-temperature
study in aqueous solution, where the solvation of fluoride ion
is particularly strong.6 Several 4-methoxybenzyl derivatives (1-
Nu) have been shown to react by concurrent concerted (ANDN)
and stepwise (DN + AN) reaction mechanisms13 through the
4-methoxybenzyl carbocation intermediate (1+) of known
lifetime,14-16 and the mechanism for reaction of 4-methoxy-
* Address correspondence to this author. E-mail: jrichard@
chem.buffalo.edu. Tel: 716 645 6800, ext. 2194. Fax: 716 645 6963.
(1) Hibbert, F.; Emsley, J. AdV. Phys. Org. Chem. 1990, 26, 255-379.
(2) Jeffrey, G. A. An Introduction to Hydrogen Bonding; Oxford University
Press: New York, 1997.
(3) Gerlt, J. A.; Gassman, P. G. J. Am. Chem. Soc. 1993, 115, 11552-11568.
(4) Guthrie, J. P. Chem. Biol. 1996, 3, 163-170.
(5) Gerlt, J. A.; Kreevoy, M. M.; Cleland, W. W.; Frey, P. A. Chem. Biol.
1997, 4, 259-267.
(6) Pearson, R. G. J. Am. Chem. Soc. 1986, 108, 6109-6114.
(7) Chapman, N. B.; Levy, J. L. J. Chem. Soc. 1952, 1677-1682.
(8) Werstiuk, N. H.; Timmins, G. Can. J. Chem. 1980, 58, 1704-1708.
(9) Swain, C. G.; Spalding, R. E. T. J. Am. Chem. Soc. 1960, 82, 6104-6107.
(10) Swain, C. G.; Knee, T. E. C.; MacLachlan, A. J. Am. Chem. Soc. 1960,
82, 6101-6104.
(12) Banait, N. S.; Jencks, W. P. J. Am. Chem. Soc. 1991, 113, 7958-7963.
(13) Guthrie, R. D.; Jencks, W. P. Acc. Chem. Res. 1989, 22, 343.
(14) Amyes, T. L.; Richard, J. P. J. Am. Chem. Soc. 1990, 112, 9507-9512.
(15) (a) Buckley, N.; Oppenheimer, N. J. J. Org. Chem. 1994, 59, 5717-5723.
(b) Buckley, N.; Oppenheimer, N. J. J. Org. Chem. 1997, 62, 540-551.
(11) Coverdale, A. K.; Kohnstam, G. J. Chem. Soc. 1960, 3806-3811.
9
9798
J. AM. CHEM. SOC. 2002, 124, 9798-9805
10.1021/ja026849s CCC: $22.00 © 2002 American Chemical Society

Hydrogen Bonding of 4-Methoxybenzyl Fluoride
A R T I C L E S
benzyl chloride has been particularly well characterized.14 It is
also possible to determine the rate constant for reaction in the
microscopic direction of the addition of halide ion to 1+, the
equilibrium constant for heterolytic bond cleavage,17,18 and the
corresponding rate and equilibrium constants for the acid-
catalyzed cleavage of 1-Nu (Scheme 1).19,20
We report here pL-rate profiles for heterolytic cleavage of
1-F in H2O and D2O, the failure to observe general acid catalysis
of this reaction, absolute rate constants for the addition of F-
and HF to the 4-methoxybenzyl carbocation (1+), and the effect
of protonation of the fluoride leaving group on the mechanism
for heterolytic C-F bond cleavage. A comparison with the
corresponding data for cleavage of 1-Cl allows an evaluation
of certain aspects of the effect of hydrogen bonding on the rate
and equilibrium constants and mechanism for heterolytic
cleavage of the carbon-fluorine bond at 1-F in water.
during the ca. 30 s required to prepare the sample for HPLC analysis.
The areas of the peaks for the products 1-F [(A1-F)t] and 1-OH
[(A1-OH)t] were obtained by HPLC analysis, and the initial areas (A1-F
)
o
and (A1-OH)o of these peaks before decay of 1-F were estimated using
eqs 1 and 2, where kobsd ) 6.6 × 10-3 s-1 is the first-order rate constant
for solvolysis of 1-F in water that contains 1 M KF, t ≈ 30 s ((1 s)
is the time between initiation of the reaction of 1-Cl and HPLC product
analysis, and 1.14 is the ratio of the extinction coefficients of 1-OH
and 1-F at 273 nm. The difference between the observed [(A1-F)t and
(A1-OH)t] and corrected [(A1-F)o and (A1-OH)o] peak areas was less than
20%.
ln(A1-F)o ) ln(A1-F)t + kobsd
t
(1)
(2)
(A1-OH)o ) (A1-OH)t - 1.14{(A1-F)o - (A1-F)t}
Kinetic Studies. The reactions of 1-F were initiated by making a
100-fold dilution of a 0.01-0.02 M solution of substrate in acetonitrile
into an aqueous solution at 25 °C and I ) 1.0 (KCl). The reaction
progress was monitored spectrophotometrically by following the change
Experimental Section
4-Methoxybenzyl fluoride (1-F) was prepared by a literature
procedure.21 4-Methoxybenzyl chloride (1-Cl) was purchased from
Aldrich and was distilled before use. A 1 M solution of LiCl was
prepared by neutralization of 1 M LiOH with HCl. All other organic
and inorganic chemicals were reagent grade and were used without
further purification. Deuterium oxide (99.9% D) was purchased from
Cambridge Isotope Laboratories.
in absorbance at 235 or 282 nm. Pseudo-first-order rate constants (kobsd
,
s-1) were determined from the slope of semilogarithmic plots of reaction
progress against time, which were linear for at least 3 halftimes. The
rate constants were reproducible to (5%.
Results
Values of ko ) 8.0 × 10-3 and 7.3 × 10-3 s-1 for the
spontaneous solvolysis of 1-F in H2O and D2O, respectively,
at 25 °C and I ) 1.0 (KCl) were determined as the observed
first-order rate constants for reaction in the presence of 0.1 M
acetate buffer ([B]/[BL+] ) 1.0). The value of ko in H2O at I )
1.0 (KCl) determined here is identical with that determined
earlier in the presence of 0.1 M NaOH at I ) 0.2 (NaClO4).24a
Specific salt effects on the solvolysis of 1-F causes ko to vary
from 8.0 × 10-3 s-1 at I ) 1.0 (KCl) to 9.3 × 10-3 and 1.3 ×
10-2 s-1, respectively, for reactions in the presence of 1.0 M
NaCl and LiCl at neutral pH. We attribute the larger 4-fold
increase to kobsd ) 3.3 × 10-2 s-1 observed for reaction in the
presence of 1.0 HCl largely or entirely to specific acid catalysis
of the solvolysis of 1-F. Values of kH ) 0.025 M-1 s-1 and kD
) 0.023 M-1 s-1 for acid-catalyzed cleavage of 1-F in H2O
and D2O, respectively, at 25 °C and I ) 1.0 (KCl) were
determined as the slope of plots of kobsd (s-1) against [H+] or
[D+] (0.01-1.0 M). The fit of the experimental data to the
logarithmic form of eq 3, obtained using these kinetic param-
eters, is shown in Figure 1 (b, H2O; 2, D2O).
Product Studies. Product studies were carried out at 25 °C and I )
1.0 (KCl). The products of the reactions of 1-Cl and 1-F were separated
and quantified by HPLC as described previously,14,22,23 except that peak
detection was by a Waters 996 diode array detector. The products were
detected by their UV absorbance at 273 nm, which is λmax for the solvent
adduct 1-OH. The ratio of the rate constants for reaction of 1-F ([S] )
0.1-0.2 mM) with solvent and azide ion in water was determined from
the ratio of the yields of 1-OH and 1-N3 as described in an earlier
study of the solvolysis of 1-Cl.14 The azide ion adduct 1-N3 was
characterized in earlier studies of nucleophilic substitution reactions at
1-X.14,15b This compound is stable to HPLC analysis: the halftime
for its solvolysis in 50/50 (v/v) trifluoroethanol/water is ca. 109-fold
longer than that (1 s) for solvolysis of 1-Cl.24 A value of ꢀ1-OH/ꢀ1-F
)
1.14 for the ratio of the extinction coefficients of 1-OH and 1-F at
273 nm was determined from the ratio of the observed peak areas from
HPLC analyses of solutions that contained known concentrations of
1-OH and 1-F.
The product rate constant ratio kF/ks (M-1) for reaction of fluoride
ion and solvent water with 1-Cl was determined from the ratio of peak
areas (A1-F)o/(A1-OH)o for the products 1-F and 1-OH formed from
reaction of 1-Cl in the presence of 1 M KF and the extinction coefficient
ratio ꢀ1-OH/ꢀ1-F ) 1.14.23 The halftime for solvolysis of 1-Cl in water
is ca. 0.2 s.24a This reaction was initiated by making a 100-fold dilution
of a solution of 1-Cl in acetonitrile into water containing 1 M KF to
give a final substrate concentration of 0.04 mM, with vigorous agitation
on a vortex mixer. The product of nucleophilic substitution of fluoride
ion, 1-F, undergoes solvolysis to give 1-OH with a halftime of 110 s
kobsd ) ko + kL[L+]
(3)
The value of kobsd ) 7.8 × 10-3 s-1 determined for reaction
of 1-F in the presence of 0.80 M cyanoacetate buffer at pH 1.7
([BH+]/[B] ) 4) and I ) 1.0 (KCl) is 10% smaller than kobsd
)
8.6 × 10-3 s-1, which can be calculated for reaction in the
absence of buffer at pH 1.7 using eq 3 and the experimental
values of ko and kH. There is a 20% decrease in kobsd for
solvolysis of 1-F as the concentration of added fluoride ion is
increased from zero to 1.0 M at a constant ionic strength of 1.0
(KCl). The product rate constant ratio for reaction of 1-Cl with
fluoride ion and solvent water, determined from the ratio of
the yields of 1-F and 1-OH from reaction of 1-Cl in the presence
of 1.0 M KF, is kF/ks ) 0.070 M-1. The value of kobsd (s-1) for
the spontaneous solvolysis of 1-F in water at I ) 1.0 (KCl)
remains constant as the concentration of azide ion is increased
(16) Kevill, D. N.; Ismail, N. H. J.; D’Souza, M. J. J. Org. Chem. 1994, 59,
6303-6312.
(17) Richard, J. P. Tetrahedron 1995, 51, 1535-1573.
(18) Richard, J. P.; Amyes, T. L.; Toteva, M. M. Acc. Chem. Res. 2001, 34,
981-988.
(19) Jensen, J. L.; Jencks, W. P. J. Am. Chem. Soc. 1979, 101, 1476-1488.
(20) Richard, J. P.; Nagorski, R. W.; Rudich, S.; Amyes, T. L.; Katritzky, A.
R.; Wells, A. P. J. Org. Chem. 1995, 60, 5989-5991.
(21) Shimizu, M.; Nakahara, Y.; Yoshioka, H. Tetrahedron Lett. 1985, 26,
4207-4210.
(22) Richard, J. P. J. Am. Chem. Soc. 1989, 111, 1455-1465.
(23) Richard, J. P.; Rothenberg, M. E.; Jencks, W. P. J. Am. Chem. Soc. 1984,
106, 1361-1372.
(24) (a) Richard, J. P.; Amyes, T. L.; Rice, D. J. J. Am. Chem. Soc. 1993, 115,
2523-2524. (b) Amyes, T. L.; Stevens, I. W.; Richard, J. P. J. Org. Chem.
1993, 58, 6057-6066.
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9799

A R T I C L E S
Toteva and Richard
Scheme 2
can be little trapping of the 4-methoxybenzyl carbocation
intermediate 1+ by fluoride ion in water. The product rate
constant ratio kF/ks ) 0.070 M-1 for reaction of 1+ with fluoride
ion and solvent water, determined from the low yield of 1-F
from the reaction of 1-Cl in the presence of 1.0 M fluoride ion,
shows that about 35% of the small observed effect of fluoride
ion on kobsd for 1-F is due to common ion inhibition; we attribute
the remainder of this effect to a specific fluoride ion salt effect
Figure 1. Dependence of log kobsd (s-1) for the solvolysis of 1-F in H2O
(b) and D2O (2) at 25 °C and I ) 1.0 (KCl) on -log [L+]. The values of
-log [H+] ) 4.52 and -log [D+] ) 5.05 were calculated from the measured
values of pH ) 4.58 and pD ) 5.11 and the relationship pL ) -log{γ[L+]},
where γ ) 0.87 is the apparent activity coefficient of hydronium ion
determined for the electrode under our experimental conditions (I ) 1.0,
KCl). The labels show (1) -log [H+] ) pKa for HF (2.9) and (2) -log
[H+] ) pHI ) 0.49, where identical first-order rate constants are observed
for the spontaneous and specific-acid-catalyzed reactions of 1-F.
on kobsd
.
The reaction of azide ion with 1+ is diffusion-limited with
kaz ) 5 × 109 M-1 s-1, so that this reaction can be used as a
“clock” to determine absolute rate constants for addition of other
nucleophiles to 1+.18,25-27 Combining the values of kaz/ks ) 20
M-1 and kF/ks ) 0.070 M-1 with kaz ) 5 × 109 M-1 s-1 gives
absolute rate constants ks ) 2.5 × 108 s-1 and kF ) 1.8 × 107
M-1 s-1 (Table 1) for addition of water and fluoride ion,
respectively, to 1+.
pL-Rate Profiles for Cleavage of 1-F. The pL-rate profiles
for cleavage of 1-F in Figure 1 show the relative rates of the
spontaneous (ko) and specific-acid-catalyzed (kH) cleavage
reactions of 1-F to form the carbocation intermediate 1+. It is
less well recognized that such pH-rate profiles also define the
relative rates of the “uncatalyzed” and “hydrogen-ion-catalyzed”
reactions in the microscopic reaction, which in this case
correspond to the addition of F- (kF) and HF (kHF) to the
4-methoxybenzyl carbocation (Scheme 3).
from zero to 0.10 M at a constant ionic strength of 1.0 (KCl).
The product rate constant ratio kaz/ks ) 20 M-1 ((10%) for
reaction of 1-F with azide ion and solvent water is the average
of the rate constant ratios determined at five different concentra-
tions of azide ion between zero and 0.1 M.
Discussion
Reaction Mechanism. The observed first-order rate constant
for reaction of 1-F in unbuffered solution remains constant as
the concentration of azide ion is increased from zero to 0.1 M
at 25 °C (I ) 1.0, KCl): these reactions give up to a 67% yield
of the azide ion adduct 1-N3. This result shows that, in water,
1-F reacts with nucleophiles by the same stepwise DN + AN
(SN1)13 mechanism through the 4-methoxybenzyl carbocation
1+ that is observed for other 4-methoxybenzyl derivatives
(Scheme 2).14 Competing stepwise and concerted nucleophilic
substitution by azide ion has been reported for 1-Cl14 and
+ 15,16
1-SR2
under some reaction conditions. However, we
The shape of the pL-rate profiles in Figure 1 is defined by
the value of -log[L+], at which identical first-order rate
constants are observed for the uncatalyzed and acid-catalyzed
reactions of 1-F, -log[H+]I ) pHI.19 Equation 5 gives the
relationship between pHI, the pKa of the conjugate acid of the
leaving group anion, and the rate constant ratio log(kF/kHF) for
the reaction of the carbocation intermediate 1+ with F- and HF.
Equation 5 follows from eq 4, derived for Scheme 3, and the
relationship kH[H+]I ) ko, which holds when the reaction is
conducted at pHI.19 Equation 5 shows that the shape of the pL-
rate profiles for the stepwise solvolysis of 1-F is controlled by
the rate constant ratio for partitioning of the reaction intermedi-
ate, log(kF/kHF), and the pKa of the conjugate acid of the fluoride
leaving group (HF). The maximum possible pHI when the
leaving group is fluoride ion is the pKa of 2.9 for HF:28a at this
maximum, log(kF/kHF) ) 0 (eq 5), and there is no selectivity
for the addition of fluoride ion and neutral HF to 1+. Values of
observe no significant bimolecular substitution reaction of 1-F
with azide ion in water.
The value of kaz/ks ) 20 M-1 for partitioning of 1+ between
addition of azide ion and solvent in water (I ) 1.0, KCl)
(Scheme 2) is similar to the value of kaz/ks ) 25 M-1 for
partitioning of 1+ in 50/50 (v/v) water/trifluoroethanol (I ) 0.50,
NaClO4) determined in earlier work.14
kF
ko
)
(4)
(5)
kHF kHKa
kF
pHI ) pKa - log
(
)
kHF
It has been reported in earlier work that the effect of similar
changes in solvent and salt on kaz/ks (M-1) for partitioning of
the 1-(4-methoxyphenyl)ethyl carbocation is ∼2-fold.23
The 20% decrease in kobsd for solvolysis of 1-F in water as
the concentration of added fluoride ion is increased from zero
to 1.0 M at I ) 1.0 (KCl) shows that there is little common ion
inhibition of the solvolysis of 1-F in water. Therefore, there
(25) McClelland, R. A.; Kanagasabapathy, V. M.; Steenken, S. J. Am. Chem.
Soc. 1988, 110, 6913-6914.
(26) McClelland, R. A.; Kanagasabapathy, V. M.; Banait, N. S.; Steenken, S.
J. Am. Chem. Soc. 1991, 113, 1009-1014.
(27) McClelland, R. A.; Cozens, F. L.; Steenken, S.; Amyes, T. L.; Richard, J.
P. J. Chem. Soc., Perkin Trans. 2 1993, 1717-1722.
9
9800 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

Hydrogen Bonding of 4-Methoxybenzyl Fluoride
A R T I C L E S
Table 1. Rate and Equilibrium Constants for Reactions of 4-Methoxybenzyl Derivatives in Water at 25 °C
Nu-
kNu
(M-1 s-1
kob
K
Nu (M-1 c
)
kHNu
(M-1 s-1
kHe
a
d
(pK )
a
)
(s-1
)
(log KNu)
)
(M-1 s-1
)
log K
T
log KXg
f
F-
1.8 × 107h
2 × 109i
8.0 × 10-3h
4j
2.3 × 109
7.2 × 104h
2.5 × 10-2h
6.5
0
(2.9)
(9.4)
Cl-
5.0 × 108
(8.7)
16.7k
10.6
10.2
4.1
(-8)
HO-
(15.7)
2 × 1026k
4.5 × 106h
1.1 × 10-4l
(26.3)
a Second-order rate constant for addition of the anionic nucleophile Nu- to 1+. b First-order rate constant for heterolytic cleavage of 1-Nu to give 1+ and
Nu-. c Equilibrium constant for addition of the anionic nucleophile Nu- to 1+, calculated as KNu ) kNu/ko, unless noted otherwise (Scheme 7). d Second-
order rate constant for addition of the protonated nucleophile HNu to 1+. e Second-order rate constant for acid-catalyzed cleavage of 1-Nu to give 1+ and
HNu. f Equilibrium constant for transfer of Nu- from the proton to 1+, calculated as KT ) kHNu/kH, unless noted otherwise (Scheme 7). g Equilibrium
constant for the isodesmic exchange of Nu- and F- between the proton and 1+ (Scheme 8). h Data from this work at I ) 1.0 (KCl). i In 50/50 (v/v) water/
trifluoroethanol at I ) 0.50 (NaClO4) [ref 31a]. j Data from ref 24a. k Calculated using the relationship pKa ) log KNu - log KT (Scheme 7). l Ref 59.
Scheme 3
unstable and the leaving group/nucleophile anion is very
reactive. For example, the value of pKa - pHI ) log(kRS/kRSH
)
) 1.6 determined for C-S bond cleavage at benzaldehyde
O-ethyl,S-phenyl acetal (Scheme 4) is consistent with solvolysis
through an oxocarbenium ion intermediate that shows a small
selectivity between the diffusion-limited addition of thio-
phenoxide ion (kPhS ) 5 × 109 M-1 s-1) and the addition of
neutral thiophenol (kPhSH ≈ 1 × 108 M-1 s-1).19 At the other
extreme, the rate constant ratio kRS/kRSH ) 2 × 107 determined
for the addition of HOCH2CH2S- and HOCH2CH2SH to the
tris(4-methoxyphenyl)methyl carbocation corresponds to a 10.0
kcal/mol difference in the barriers to the reaction of these
nucleophiles at pH ) pKa for mercaptoethanol.29
pHI > pKa are not expected, because this would require log(kF/
The difference in the selectivities for addition of fluorine
nucleophiles to 1+ (kF/kHF ) 250) and sulfur nucleophiles to
the tris(4-methoxyphenyl)methyl carbocation (kRS/kRSH ) 2 ×
107) may simply reflect a smaller nucleophilic selectivity for
nucleophile addition to the more reactive carbocation 1+.
However, the magnitude of kNu/kHNu for reaction of an anion
and its conjugate acid in water depends on both the relative
intrinsic reactivity of these nucleophiles and the attenuation of
this reactivity by aqueous solvation. The strong solvation of
the fluorine nucleophiles F- and HF probably reduces the value
of kF/kHF relative to kRS/kRSH, because of the tendency of
solvation either to decrease the reactivity of fluoride ion or
increase the reactivity of its conjugate acid (Figure 2A).
The strong solvation of fluoride ion in water6 and the effect
of this solvation on the nucleophilic reactivity of fluoride ion
in concerted bimolecular substitution at aliphatic carbon has been
extensively documented.30 The data reported here provide
evidence for a large effect of the strong aqueous solvation of
fluoride ion on the reactivity of this nucleophile in carbocation
addition (Figure 2A). For example, the value of kF ) 1.8 × 107
M-1 s-1 for addition of F- to 1+ is ca. 100-fold smaller than
kCl ) 2 × 109 M-1 s-1 for addition of the much more weakly
basic chloride ion,31a and only 3-fold larger than kHOH ) 4.5 ×
106 M-1 s-1 for addition of water (Table 1).31b The low 7%
yield of 1-F from partitioning of 1+ generated from reaction of
k
HF) < 0 and a greater nucleophilic reactivity of HF than of the
more basic fluoride ion toward 1+.
Identical first-order rate constants are observed for the
uncatalyzed and the specific-acid-catalyzed cleavage of 1-F in
H2O at pHI ) 0.49 (Figure 1). A value of kF/kHF ) 250 was
calculated from eq 4 derived for Scheme 3, using ko/kH
)
(0.0080 s-1/0.025 M-1 s-1) and Ka ) 1.26 × 10-3 M for
ionization of HF.28a Combining this ratio with kF ) 1.8 × 107
M-1 s-1 (Table 1) gives kHF ) 7.2 × 104 M-1 s-1 for
nucleophilic addition of HF to 1+ (Table 1). Values of pHI )
1.8 and 2.4, respectively, can be calculated from earlier data
for the spontaneous and acid-catalyzed cleavage of benzyl
fluoride in 90/10 water/acetone at 50 °C9 and for the cleavage
of triphenylmethyl fluoride in 70/30 water/acetone at 40 °C.11
A quantitative analysis of these data using eq 5 is not possible,
because of uncertainties about the effect of mixed aqueous
organic solvents and the change in reaction temperature on the
pKa of HF.
Nucleophilic Reactivity of F- and HF. There is a 3.3 kcal/
mol difference in the apparent barriers for addition of anionic
F- and neutral HF to 1+ (kF/kHF ) 250, Table 1) for reactions
at pH ) pKa ) 2.9 for HF (Figure 2A). These apparent barriers
will be equal in the limiting case where kNu ) kHNu for reaction
of anionic Nu- and neutral HNu with 1+, so that {pKa - pHI}
) log(kNu/kHNu) ) 0. This limit is approached for stepwise
solvolysis reactions when the carbocation intermediate is very
(29) Ritchie, C. D.; VanVerth, J. E.; Virtanen, P. O. I. J. Am. Chem. Soc. 1982,
104, 3491-3497.
(30) Parker, A. J. Chem. ReV. 1969, 69, 1-32.
(28) (a) The dissociation constant of HF at I ) 1.0 (KCl) has not been reported.
Values of the pKa of HF at I ) 1.0 (NaClO4) ranging from 2.90 to 2.95
have been reported [Stability Constants of Metal-Ion Complexes. Special
Publication No. 17.; Chemical Society: London, 1964]. A value of pKa )
2.82 for ionization of HF in the presence of an unspecified electrolyte (I )
1.0) has been reported [Mosha, D.; Qulwi, Q.; Mhinzi, G. J. Chem. Soc.
Pak. 1996, 18, 96-100]. (b) Perrin, D. D. Dissociation Constants of
Inorganic Acids and Bases in Aqueous Solution; Butterworth: London,
1969.
(31) (a) A value of kCl ) 2 × 109 M-1 s-1 has been reported for addition of
chloride ion to 1+ in 50/50 (v/v) trifluoroethanol/water [ref 14]. It has been
shown that there is no effect of the change from this solvent to water on
the rate constant kaz ) 5 × 109 M-1 s-1 for the diffusion-limited reaction
of azide ion [refs 26 and 27]. (b) The value of kHOH ) 4.5 × 106 M-1 s-1
for the second-order rate constant for addition of water to 1+ was calculated
from kaz/ks ) 20 M-1 for partitioning of 1+ between addition of azide ion
and water, kaz ) 5 × 109 M-1 s-1 for the diffusion-limited reaction of
azide ion [refs 26 and 27], and the 55 M concentration of solvent water.
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9801

A R T I C L E S
Toteva and Richard
Figure 2. Free energy profiles that compare the activation barriers and changes in Gibbs free energy for the following reactions in water at 25 °C. (A) The
addition of F- and HF to the 4-methoxybenzyl carbocation 1+ at pH ) pKa ) 2.9 for HF. The arrows indicate the tendency of the strong solvation of
fluoride ion and hydrofluoric acid to decrease and increase, respectively, the rate constant for their reaction with 1+. (B) The heterolytic cleavage of 1-F and
1-Cl to form 1+ and the corresponding halide ion.
Scheme 4
in a concerted pathway, which has the effect of increasing kHF
(see below).
By comparison, thiol anions are relatively weakly solvated
in water6 and thiols form weak hydrogen bonds to this solvent.33
This should result in a smaller attenuation by solvation,
compared with the strongly solvated fluoride ion, of the large
difference in the intrinsic nucleophilic reactivity of these neutral
and anionic nucleophiles.
Intrinsic Reaction Barriers. Figure 2B shows the 3.7 kcal/
mol difference in the barriers to heterolytic cleavage of 1-F (ko
) 8.0 × 10-3 s-1) and 1-Cl (ko ) 4 s-1 24a
)
to give the common
carbocation intermediate 1+ in water. This difference in ko is
due primarily to the larger Marcus intrinsic barrier to the reaction
of 1-F,34 since the heterolytic cleavage of 1-F and 1-Cl are each
unfavorable by ca. 12 kcal/mol (Table 1).
1-Cl in water in the presence of 1 M fluoride ion shows that
the barrier to cleavage of a single hydrogen bond between F-
and water, which is required to free an electron pair at F- in
order for it to react with 1+, is comparable to the barrier for the
direct addition of 1 M water, so that the reaction of solvent
dominates because of its large (55 M) concentration. It is as
though the strong network of hydrogen bonds to fluoride ion
has inhibited its nucleophilic addition to 1+ by entrapping this
anion in the surrounding solvent. The relatively large nucleo-
philic selectivity observed for reaction of fluoride ion with the
covalent R-mannosyl intermediate of the reaction catalyzed by
mutant forms of a â-mannosidase provides evidence that there
is a large increase in the nucleophilic reactivity of fluoride ion
upon its transfer from water to this enzyme catalyst.32
The free energy of solvation of fluoride ion in water (ca. 105
kcal/mol) is much larger than that of chloride ion (ca. 75 kcal/
mol)6 and strongly favors processes such as heterolytic cleavage
that generate fluoride ion as a product. By contrast, the very
similar thermodynamic barrier observed for heterolytic cleavage
of 1-F and 1-Cl in water (Figure 2B and Table 1) shows that
the stronger solvation of F- is offset by the larger carbon basicity
of F- than of Cl-, which stabilizes the reactant 1-F compared
with 1-Cl. The estimated Marcus intrinsic barrier to heterolytic
cleavage of 1-F is 3 kcal/mol larger than that for cleavage of
1-Cl.35 Similar intrinsic barriers for these cleavage reactions
would be observed if the progress in the transition state toward
heterolytic cleavage of the carbon-halogen bond, which is
stronger for 1-F than 1-Cl (see above), were balanced by the
In addition, the relatively small rate constant ratio kF/kHF may
reflect stabilization of the transition state for addition of neutral
HF to 1+ by partial proton transfer to a hydrogen-bonded water
(32) Zechel, D. L.; Reid, S. P.; Nashiru, O.; Mayer, C.; Stoll, D.; Jakeman, D.
L.; Warren, R. A. J.; Withers, S. G. J. Am. Chem. Soc. 2001, 123, 4350-
4351.
(33) Jencks, W. P.; Salvesen, K. J. Am. Chem. Soc. 1971, 93, 4433-4436.
(34) Marcus, R. A. J. Chem. Phys. 1956, 24, 966-978. Marcus, R. A. J. Phys.
Chem. 1968, 72, 891-899.
9
9802 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

Hydrogen Bonding of 4-Methoxybenzyl Fluoride
A R T I C L E S
Scheme 5
deuterium isotope effects kH/kD ) 0.33-0.50 observed for
stepwise acid-catalyzed solvolysis reactions.43 Such inverse
solvent deuterium isotope effects are a consequence of the larger
concentration of the protonated substrate in D2O, which arises
from the greater tendency of D3O+ than of H3O+ to undergo
proton transfer to form a monoprotic acid and a molecule of
L2O.43 The solvent deuterium isotope effect for acid-catalyzed
cleavage of 1-F of near unity observed here is consistent with
a concerted reaction mechanism (kcon, Scheme 5A) in which
there is movement of the proton from L3O+ to the leaving group
in the rate-determining transition state, so that the 3-fold greater
acidity of D3O+ compared with H3O+ is balanced by a normal
primary isotope effect.
There is good evidence that a concerted mechanism for acid-
catalyzed heterolytic bond cleavage at carbon is favored when
the leaving group is weakly basic,44-49 because a decrease in
leaving group basicity is accompanied by a decrease in the
stability of the protonated intermediate of the stepwise reaction
(e.g., 1-FH+), and this will favor the concerted mechanism that
avoids its formation. For example, the increase in the solvent
deuterium isotope effect from kH/kD ) 0.33 to kH/kD ) 1.09
for acid-catalyzed cleavage of benzaldehyde mixed acetals
PhCH(OMe)OR that is observed as the leaving group ROH is
changed from methanol (pKa ) 15.5) to m-nitrophenol (pKa )
8.4) is consistent with a change from a stepwise to a concerted
reaction mechanism.47 This change in mechanism has been
thoroughly documented for general acid catalysis of the cleavage
of acetals45-47 and, in the microscopic direction, for general
base catalysis of the addition of alcohols to ring-substituted
1-phenylethyl carbocations.48,49 By contrast, it is less commonly
observed for specific acid catalysis of heterolytic bond cleav-
age.46
progress toward development of the stabilizing aqueous solva-
tion of the developing product anion, which is more stabilizing
for fluoride than for chloride ion. The estimated 3 kcal/mol
larger intrinsic barrier to heterolytic cleavage of 1-F compared
with 1-Cl in water is consistent with an imbalance between the
fraction of cleavage of the carbon-halogen bond (large) and
the fractional development of the total anion stabilization by
solvation in the transition state for heterolytic C-X bond
cleavage (relatively smaller). This results in a larger barrier to
formation of the transition state, where the requirement for
stabilization by solvation is large.36-38
The data reported here are consistent with the notion that
there is little or no solvation of the electron pair used in
formation of the C-Cl and C-F bonds at the transition state
for heterolytic C-X bond cleavage to form the intimate ion
pair reaction intermediate, so that this solvation develops only
after the transition state for cleavage of 1-Nu, when the intimate
ion pair undergoes separation to the solvent-separated ion
pair.39-41
Catalysis by the Proton. The large reactivity of HF (pKa ≈
-12 for H2F+)42a compared to much more basic F- (pKa ) 2.9
for HF)28a toward 1+ in water may reflect the advantage of a
concerted compared with a stepwise mechanism for the acid-
catalyzed cleavage and synthesis of the C-F bond (kcon, Scheme
5A). A concerted mechanism would result in an increase in the
nucleophilicity of HF due to partial proton transfer to the
hydrogen bond acceptor water in the transition state for
nucleophilic addition of HF. The solvent deuterium isotope
effect kH/kD ) 1.09 (Results section) for acid-catalyzed cleavage
of 1-F determined here is larger than the inverse solvent
The substrate 1-F is so weakly basic that its protonation by
H3O+ to form solvent-equilibrated 1-FH+ (Scheme 5A) is not
expected to be much faster than the overall rate of its acid-
catalyzed cleavage to give 1+ and HF (kH ) 0.025 M-1 s-1
,
Table 1). For example, a value of kp ) 0.1 M-1 s-1 for formation
of solvent-equilibrated 1-FH+ can be calculated with the limiting
value of k-p ) kreorg ≈ 1011 s-1 for deprotonation of 1-FH+
(Scheme 5B, reverse reaction),50 and 1/Ka ) kp/k-p ) 10-12
for this protonated alkyl fluoride.42b This calculation is ap-
proximate, because the acidity of protonated alkyl fluorides is
poorly documented, but it does suggest that solvent-equilibrated
1-FH+ forms only barely fast enough for it to be an intermediate
in the acid-catalyzed cleavage of 1-F.
(35) Intrinsic barriers of Λ ) 9.8 and Λ ) 13.2 kcal/mol for heterolytic cleavage
of 1-Cl and 1-F, respectively, can be calculated using eq 6 (derived at 298
K) using ko ) 4 s-1 and K ) 1/KNu ) 2 × 10-9 M for the reaction of 1-Cl,
and ko ) 0.0080 s-1 and K ) 1/KNu ) 4.3 × 10-10 M for the reaction of
1-F.
It is interesting to consider whether the concerted mechanism
for acid-catalyzed cleavage of 1-F is enforced because the
putative intermediate 1-FH+ of the stepwise reaction is too
unstable to exist in a potential energy well for the time of a
bond vibration (Scheme 5A).51,52 There is in fact little or no
2
1
1.36
1.36 log K
4Λ
log ko )
17.44 - Λ 1 -
(6)
{
(
) }
(36) Bernasconi, C. F. Acc. Chem. Res. 1992, 25, 9-16.
(37) Bernasconi, C. F. AdV. Phys. Org. Chem. 1992, 27, 119-238.
(38) Bernasconi, C. F. Acc. Chem. Res. 1987, 20, 301-308.
(39) Richard, J. P. J. Chem. Soc., Chem. Commun. 1987, 1768-1769.
(40) Richard, J. P.; Jencks, W. P. J. Am. Chem. Soc. 1984, 106, 1373-1383.
(41) McClelland, R. A.; Kanagasabapathy, V. M.; Banait, N. S.; Steenken, S.
J. Am. Chem. Soc. 1992, 114, 1816-1823.
(43) Schowen, R. L. Prog. Phys. Org. Chem. 1972, 9, 275-332.
(44) Jencks, W. P. Acc. Chem. Res. 1976, 9, 425-432.
(45) Fife, T. H. AdV. Phys. Org. Chem. 1975, 11, 1-122.
(42) (a) The pKa of the very strong acid H2F+ in water has not been determined.
A pKa of -12 can be estimated from Ks ) 10-12 M2 for the autoionization
constant of pure (50 M) hydrofluoric acid [Gillespie, R. J.; Liang, J. J.
Am. Chem. Soc. 1988, 110, 6053-6057] and the pKa of 3.2 for the ionization
of HF in water [ref 28b], with the assumption that the equilibrium constant
for proton transfer between two HF molecules to form H2F+ and F- in
water is the same as that in pure HF. (b) This pKa is estimated to be similar
to the pKa of -12 estimated for H2F+ [ref 42a] based on the similar pKa’s
(46) Capon, B.; Nimmo, K. J. Chem. Soc., Perkin Trans. 2 1975, 1113-1118.
(47) Jensen, J. L.; Herold, L. R.; Lenz, P. A.; Trusty, S.; Sergi, V.; Bell, K.;
Rogers, P. J. Am. Chem. Soc. 1979, 101, 4672-4677.
(48) Richard, J. P.; Jencks, W. P. J. Am. Chem. Soc. 1984, 106, 1396-1401.
(49) Ta-Shma, R.; Jencks, W. P. J. Am. Chem. Soc. 1986, 108, 8040-8050.
(50) A value of kreorg ) 1011 s-1 is used for this limiting rate constant: this is
the rate constant for reorganization of the solvent by dielectric relaxation
[Giese, K.; Kaatze, U.; Pottel, R. J. Phys. Chem. 1970, 74, 3718-3725.
Kaatze, U. J. Chem. Eng. Data 1989, 34, 371-374. Kaatze, U.; Pottel, R.;
Schumacher, A. J. Phys. Chem. 1992, 96, 6017-6020].
+
+
for NH4 (pKa ) 9.25) and PhCH2NH3 (pKa ) 9.34) [Jencks, W. P.;
Regenstein, J. In Handbook of Biochemistry and Molecular Biology,
Physical and Chemical Data; 3rd ed.; Fasman, G. D., Ed.; CRC Press:
Cleveland, OH, 1976; Vol. 1, pp 305-351].
(51) Jencks, W. P. Acc. Chem. Res. 1980, 13, 161-169.
(52) Jencks, W. P. Chem. Soc. ReV. 1981, 10, 345-375.
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9803

A R T I C L E S
Toteva and Richard
Scheme 6
Figure 3. Free energy profile for the specific-acid-catalyzed cleavage of
1-F by a hypothetical stepwise reaction mechanism (Scheme 5). The arrow
is included to indicate that this reaction does in fact proceed by a concerted
mechanism, which avoids formation of the highly unstable protonated alkyl
fluoride 1-FH+.
) 3.2 for HF)28b (“libido” rule of Jencks).55 Catalysis of the
cleavage of alkyl fluorides by cyanoacetic acid (pKa ) 2.2) is
allowed by this libido rule,55 because proton transfer from
cyanoacetic acid to fluoride ion is just favorable thermodynami-
cally. However, there is no increase in the observed rate constant
for solvolysis of 1-F in the presence of 0.80 M cyanoacetic
acid buffer at pH 1.7. We conclude that, under these reaction
conditions, there is little or no stabilization of the transition state
of the specific-acid-catalyzed reaction of 1-F by replacement
of the hydrogen bond to solvent water in the transition state
with a hydrogen bond to cyanoacetate anion (Scheme 6).
The failure to observe general acid catalysis of the cleavage
of 1-F in water by cyanoacetic acid is consistent with the
conclusion that water (pKa ) -1.8 for H3O+) and cyanoacetate
anion form hydrogen bonds of comparable stability in the
transition state, but that the solvent reaction dominates because
of the large (55 M) concentration of water.44 The catalysis by
chloroacetic acid that has been reported for the cleavage of
benzaldehyde methyl phenyl acetal reflects the large negative
deviation of the rate constant for the solvent (hydronium ion)-
catalyzed reaction from the Brønsted plot for general acid
catalysis.44 By analogy, the absence of significant general acid
catalysis of cleavage of 1-F by cyanoacetic acid requires that
the rate constant for hydronium ion show a smaller negative
deviation from the corresponding Brønsted correlation for this
reaction. We do not understand why the relative effectiveness
of hydronium ion and carboxylic acids as catalysts should be
different for different reactions. Finally, catalysis of the cleavage
of trityl fluoride by HF has been reported for reaction in a mixed
aqueous/organic solvent.11 Such catalysis probably reflects the
well-documented strength of the hydrogen bond between
fluoride ion and HF in the complex [F‚H‚F]-.1
Carbon and Proton Basicity. Table 1 summarizes the rate
and equilibrium constants for the reactions in Schemes 7 and 8
for Nu- ) F- (this work), Cl-, and HO-. The equilibrium
constants for ionization of HNu (Ka, M) and for addition of
Nu- to 1+ (KNu ) kNu/ko, M-1) were combined to give values
of the overall equilibrium constant KT for transfer of anions
from the proton to 1+ (KT ) KNuKa, Scheme 7). The values of
log KT increase from 6.5 to 16.7 as the nucleophile/leaving group
is changed from fluoride to chloride anion. Combining these
values of KT for reactions of different anions gives values of
KX for the isodesmic exchange of Nu- and F- between the
proton and 1+ (log KX ) log [(KT)X/(KT)F], Scheme 8). The
values of KX increase from 0 (by definition) for Nu- ) F- to
10.6 for Nu- ) Cl-.
barrier to proton transfer from 1-FH+ to water (reverse of
reaction shown in Scheme 5B), because of the large thermo-
dynamic driving force for this reaction (pKa ≈ -12 for
1-FH+)42b and the small intrinsic barrier to proton transfer
between electronegative atoms.53 Therefore, the rate-determining
step for the deprotonation of 1-FH+ by water to form 1-F
(Scheme 5B, reverse reaction) and, by microscopic reversibility,
for proton transfer from H3O+ to 1-F should be the “rotation”
of a molecule of solvent water into or out of a “reactive” position
at 1-FH+ (kreorg ≈ 1011 s-1),50 so that k-p′ . kreorg (Scheme
5B). There is, however, likely a small barrier for C-F bond
cleavage at 1-FH+ to form 1+ (ki, Scheme 5A). The alternative
“barrierless” cleavage reaction would require that the change
in leaving group from F- to HF result in a ca. 1015-fold increase
in the rate constant for cleavage of the C-F bond, from ko )
8.0 × 10-3 s-1 for cleavage of 1-F, to ki ≈ 1013 s-1 for cleavage
of 1-FH+. This is approximately the same as the ca. 1015-fold
difference in the basicities of the F- and HF leaving groups.28,42a
It would correspond to values of âlg ≈ -1.0 and ânuc ) 0, and
it would require that there be essentially no chemical selectivity
for the addition of F- and HF to 1+. However, there is good
evidence for a small, but significant, selectivity for nucleophile
addition to 1+.14
The available experimental data are consistent with the notion
that there is a “flat” potential energy surface for the hypothetical
stepwise specific-acid-catalyzed cleavage of 1-F (Figure 3). It
has been proposed that such a flat surface in the region of an
unstable intermediate such as 1-FH+ will favor the observation
of a concerted mechanism which avoids its formation.54 This
is because the barrier to a concerted reaction is related to the
sum of the barriers for the two steps in the stepwise reaction
(kp and ki, Scheme 5A), minus the advantage of the coupling
of these two processes in a concerted process.54 The favorable
advantage of coupling will most likely reduce the barrier for a
concerted reaction below that for the competing stepwise
reaction when there is a shallow potential energy well (or no
well at all) for the reaction intermediate (Figure 3).
General Acid Catalysis. Catalysis of the hydrolysis of R-D-
glucopyranosyl fluoride by phosphate and phosphonate buffers
has been reported and shown to be due mainly to the basic form
of the buffer.12 There should in fact be little or no general acid
catalysis of this reaction, because of the absence of a significant
thermodynamic driving force for proton transfer from these
buffer acids (pKa g 4.3) to the fluoride ion leaving group (pKa
The data in Table 1 are consistent with the known preference
of “soft” acids and bases (1+ and Cl-) and “hard” acids and
(53) Eigen, M. Angew. Chem., Int. Ed. Engl. 1964, 3, 1-72.
(54) Dewar, M. J. S. J. Am. Chem. Soc. 1984, 106, 209-219.
(55) Jencks, W. P. J. Am. Chem. Soc. 1972, 94, 4731-4732.
9
9804 J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002

Hydrogen Bonding of 4-Methoxybenzyl Fluoride
A R T I C L E S
Scheme 7
difference in the energies of the C-X bond at 1-F and 1-Cl to
be similar to the 24.5 kcal/mol difference in the bond dissocia-
tion energies for CH3CH2-F (106 kcal/mol) and CH3CH2-Cl
(81.5 kcal/mol).57 This attenuation of bond dissociation energies
reflects the less favorable overlap between the sp3-hybrid orbitals
of carbon compared to the 1s orbital of hydrogen with the
corresponding sp3-hybrid orbitals of chlorine and fluorine.
For reactions in water, the much larger range of the proton
basicity compared with the carbon basicity of Nu- leads directly
to the large observed changes in KT. For example, there is an
11 unit difference between the pKa’s of HF (pKa ) 3.2)28b and
HCl (pKa ≈ -8),6 but little difference in the observed carbon
basicity of F- and Cl- toward 1+ (∆log KNu ) 0.7, Table 1
and Figure 2B). This very large difference in the acidity of HF
and HCl in water arises from the ca. 38 kcal/mol larger (less
favorable) free energy change for heterolytic bond cleavage of
HF than of HCl in the gas phase58 and the 7 kcal/mol greater
(more negative) free energy of solvation of HF than of HCl,6
the sum of which is greater than the compensating ca. 30 kcal/
mol larger (more negative) free energy of solvation of the
product fluoride ion than of chloride ion.6
Scheme 8
bases (H+ and F-) to react with one another, and they can be
rationalized within the framework developed to explain this
general trend of organic and inorganic reactivity.56 There is a
good correlation for reaction of chorine, fluorine, and oxygen
nucleophiles between the electronegativity of the electron donor
atom at Nu and the equilibrium constant KX for Scheme 8. This
correlation results from (1) the large difference of 32.8 kcal/
mol in the bond dissociation energies DX for H-F (135.8 kcal/
mol) and H-Cl (103.0 kcal/mol),57 which reflects the optimal
overlap between the 1s orbital of hydrogen and the correspond-
ing sp3-hybrid orbitals of chlorine and fluorine, and (2) the
attenuation of this large difference on moving to the bond
dissociation energies for 1-F and 1-Cl. We estimate the
Acknowledgment. We acknowledge National Institutes of
Health Grant GM 39754 for generous support of this work. We
thank Tina L. Amyes for helpful discussions.
JA026849S
(58) NIST Standard Reference Database Number 69-July 2001 Release (http://
webbook.nist.gov/chemistry/).
(59) Calculated from kH ) 3.7 × 10-5 M-1 s-1 for acid-catalyzed cleavage of
1-OMe in water at I ) 0.30 (KCl)24a and a ratio of 0.11/0.038 ) 2.9 for
the rate constants for cleavage of 1-(4-methoxyphenyl)ethyl alcohol (2-
OH) and 1-(4-methoxyphenyl)ethyl methyl ether (2-OMe) in 50/50 (v/v)
trifluoroethanol at 25 °C and I ) 0.50 (NaClO4) (eq 7). An experimental
value of (kH)HO ) 0.11 M-1 s-1 (eq 7) is used [ref 23]. The value of (kH)MeO
) 0.038 M-1 s-1 for cleavage of 2-OMe was calculated from (kH)HO, Keq
) 63 for interconversion of 1-(4-methoxyphenyl)ethyl alcohol and 1-(4-
methoxyphenyl)ethyl methyl ether [Rothenberg, M. E.; Richard, J. P.; Jencks
W. P. J. Am. Chem. Soc. 1985, 107, 1340-1346], and kMeOH/kHOH ) 22
for partitioning of the 1-(4-methoxyphenyl)ethyl carbocation 2+ between
addition of methanol and water [ref 40] (eq 7).
(56) Pearson, R. G., Ed. Hard and Soft Acids and Bases; Dowden, Hutchinson
and Ross: Stroudsberg, PA, 1973. Pearson, R. G. Chemical Hardness; John
Wiley & Sons: New York, 1997.
(57) Benson, S. W. J. Chem. Educ. 1965, 42, 502-518.
+
(kH)HO[H
]
kMeOH
MeOH + 2-OH y
z 2+ y
z HOH + 2-OMe (7)
kHOH
(kH)
MeO[H+
]
9
J. AM. CHEM. SOC. VOL. 124, NO. 33, 2002 9805