Substituent effects on carbocation stability: The pKR for p-quinone methide

Article abstract of DOI:10.1021/ja029588v

A value of k_H = 1.5 × 10^-3 M^-1 s^-1 has been determined for the generation of simple p-quinone methide by the acid-catalyzed cleavage of 4-hydroxybenzyl alcohol in water at 25 °C and / = 1.0 (NaClO₄). This was combined with k_s = 5.8 × 106 s^-1 for the reverse addition of solvent water to the 4-hydroxybenzyl carbocation [J. Am. Chem. Soc. 2002, 124, 6349-6356] to give pK_R = -9.6 as the Lewis acidity constant of O-protonated p-quinone methide. Values of pK_R = 2.3 for the Lewis acidity constant of neutral p-quinone methide and pK_add = -7.6 for the overall addition of solvent water to p-quinone methide to form 4-hydroxybenzyl alcohol are also reported. The thermodynamic driving force for transfer of the elements of water from formaldehyde hydrate to p-quinone methide to form formaldehyde and p-(hydroxymethyl)-phenol (4-hydroxybenzyl alcohol) is determined as 6 kcal/mol. This relatively small driving force represents the balance between the much stronger chemical bonds to oxygen at the reactant formaldehyde hydrate than at the product p-(hydroxymethyl) phenol and the large stabilization of product arising from the aromatization that accompanies solvent addition to p-quinone methide. The Marcus intrinsic barrier for nucleophilic addition of solvent water to the "extended" carbonyl group at p-quinone methide is estimated to be 4.5 kcal/mol larger than that for the addition of water to the simple carbonyl group of formaldehyde. O-Alkylation of p-quinone methide to give the 4-methoxybenzyl carbocation and of formaldehyde to give a simple oxocarbenium ion results in very little change in the relative Marcus intrinsic barriers for the addition of solvent water to these electrophiles.

Full text of DOI:10.1021/ja029588v

Published on Web 07/01/2003
Substituent Effects on Carbocation Stability: The pKR for
p-Quinone Methide
Maria M. Toteva, Michael Moran,† Tina L. Amyes, and John P. Richard*
Contribution from the Department of Chemistry, UniVersity at Buffalo, SUNY,
Buffalo, New York 14260-3000
Received December 4, 2002; E-mail: jrichard@chem.buffalo.edu
Abstract: A value of kH ) 1.5 × 10-3 M-1 s-1 has been determined for the generation of simple p-quinone
methide by the acid-catalyzed cleavage of 4-hydroxybenzyl alcohol in water at 25 °C and I ) 1.0 (NaClO4).
This was combined with ks ) 5.8 × 106 s-1 for the reverse addition of solvent water to the 4-hydroxybenzyl
carbocation [J. Am. Chem. Soc. 2002, 124, 6349-6356] to give pKR ) -9.6 as the Lewis acidity constant
of O-protonated p-quinone methide. Values of pKR ) 2.3 for the Lewis acidity constant of neutral p-quinone
methide and pKadd ) -7.6 for the overall addition of solvent water to p-quinone methide to form
4-hydroxybenzyl alcohol are also reported. The thermodynamic driving force for transfer of the elements
of water from formaldehyde hydrate to p-quinone methide to form formaldehyde and p-(hydroxymethyl)-
phenol (4-hydroxybenzyl alcohol) is determined as 6 kcal/mol. This relatively small driving force represents
the balance between the much stronger chemical bonds to oxygen at the reactant formaldehyde hydrate
than at the product p-(hydroxymethyl)phenol and the large stabilization of product arising from the
aromatization that accompanies solvent addition to p-quinone methide. The Marcus intrinsic barrier for
nucleophilic addition of solvent water to the “extended” carbonyl group at p-quinone methide is estimated
to be 4.5 kcal/mol larger than that for the addition of water to the simple carbonyl group of formaldehyde.
O-Alkylation of p-quinone methide to give the 4-methoxybenzyl carbocation and of formaldehyde to give a
simple oxocarbenium ion results in very little change in the relative Marcus intrinsic barriers for the addition
of solvent water to these electrophiles.
Scheme 1
Introduction
The parent p-quinone methide 1 and its relatives that contain
the quinone methide functionality have long attracted the interest
of discerning chemists.1-11 1 can be thought of as a formally
neutral benzylic carbocation at which there is limiting resonance
stabilization by electron donation from a p-oxygen anion sub-
stituent to the cationic benzylic carbon.12-14 This strong inter-
action results in a high kinetic stability of, and large nucleophile
selectivities toward, quinone methides13,14 and is responsible
in part for the interesting biological activity observed for more
complex quinone methides.15-20 The key property that distin-
guishes the 1,6-addition of nucleophiles to p-quinone methides
from related Michael addition reactions is the aromatization of
the formal cyclohexadiene ring that accompanies this addition
(Scheme 1). We are interested in evaluating the effect of this
aromatic ring formation on the thermodynamic driving force
for 1,6-addition of HNu to 1 through a comparison of this
reaction with the related 1,2-addition of HNu to formaldehyde
(Scheme 1).
* To whom correspondence should be addressed. Phone: 716-645-6800
ext 2194. Fax: 716-645-6963.
† Deceased.
(1) Pande, P.; Shearer, J.; Yang, J.; Greenberg, W.; Rokita, S. E. J. Am. Chem.
Soc. 1999, 121, 6773-6779.
(2) Filar, L. J.; Winstein, S. Tetrahedron Lett. 1960, 25, 9-16.
(3) Fan, P. W.; Zhang, F.; Bolton, J. L. Chem. Res. Toxicol. 2000, 13, 45-52.
(4) Chiang, Y.; Kresge, A. J.; Zhu, Y. J. Am. Chem. Soc. 2000, 122, 9854-
9855.
(5) de la Mare, P. B. D.; Newman, P. A. J. Chem. Soc., Perkin Trans. 2 1984,
231-238.
(6) Wan, P.; Barker, B.; Diao, L.; Fisher, M.; Shi, Y.; Yang, C. Can. J. Chem.
1996, 74, 465-475.
The photochemical generation of simple quinone methides
was first reported by Wan and co-workers.21 This was followed
by the determination by Kresge and co-workers of rate constants
(7) Zhou, Q.; Turnbull, K. J. Org. Chem. 1999, 64, 2847-2851.
(8) Stowell, J. K.; Widlanski, T. S.; Kutateladze, T. G.; Raines, R. T. J. Org.
Chem. 1995, 60, 6930-6936.
(9) Freccero, M.; Valentin, C. D.; Sarzi-Amade, M. J. Am. Chem. Soc. 2003,
125, 3544-3553.
(10) Burnham, K. S.; Schuster, G. B. J. Am. Chem. Soc. 1998, 120, 12619-
12625.
(15) Boldt, M.; Gaudiano, G.; Koch, T. H. J. Org. Chem. 1987, 52, 2146-
2153.
(11) Rabin, O.; Vigalok, A.; Milstein, D. J. Am. Chem. Soc. 1998, 120, 7119-
(16) Fisher, J.; Ramakrishnan, K.; Becvar, J. E. Biochemistry 1983, 22, 1347-
7120.
1355.
(12) Richard, J. P.; Amyes, T. L.; Toteva, M. M. Acc. Chem. Res. 2001, 34,
981-988.
(17) Gaudiano, G.; Frigerio, M.; Sangsurasak, C.; Bravo, P.; Koch, T. H. J.
Am. Chem. Soc. 1992, 114, 5546-5553.
(18) Kujmar, S.; Lipmna, R.; Tomasz, M. Biochemistry 1992, 31, 1399-1407.
(19) Han, I.; Russell, D. J.; Kohn, H. J. Org. Chem. 1992, 57, 1799-1807.
(20) Boruah, R. C.; Skibo, E. B. J. Org. Chem. 1995, 60, 2232-2243.
(13) Richard, J. P.; Toteva, M. M.; Crugeiras, J. J. Am. Chem. Soc. 2000, 122,
1664-1674.
(14) Toteva, M. M.; Richard, J. P. J. Am. Chem. Soc. 2000, 122, 11073-11083.
9
8814
J. AM. CHEM. SOC. 2003, 125, 8814-8819
10.1021/ja029588v CCC: $25.00 © 2003 American Chemical Society

Richard et al.
A R T I C L E S
for nucleophilic addition of solvent water to p-quinone methide
1, protonated p-quinone methide H-1+, and their o-quinone
methide counterparts, generated by laser flash photolysis.4,22-24
There have also been several studies of the generation of quinone
methides as intermediates of solvolysis reactions.2,5,13,14,25
Kinetic data for the acid-catalyzed cleavage of 4-hydroxybenzyl
alcohol H-1-OH would be of particular interest, because they
represent
the
“missing
link”
in
the
exper-
imental determination of the Lewis acidity constants for the
addition of solvent water to 1 and H-1+. These acidity constants
are required to establish the electrophilicity of simple quinone
methides relative to that of other prototypical electrophiles such
as the 4-methoxybenzyl carbocation Me-1+ and the simple
carbonyl compound formaldehyde.
Figure 1. (A) Dependence of the observed first-order rate constant for
acid-catalyzed nucleophilic substitution of thiol at H-1-OH to form H-1-
SCH2CH2OH on the concentration of 2-mercaptoethanol in water at
25 °C and I ) 1.0 (NaClO4): (b) [HClO4] ) 1.00 M; ([) [HClO4] ) 0.80
M; (1) [HClO4] ) 0.45 M; (9) [HClO4] ) 0.094 M. Solid lines through
the data were calculated using eq 1 with the values of klim (s-1) and kRSH/ks
(M-1) reported in Table 1. (B) Dependence of the limiting first-order rate
constant klim from Figure 1A on the concentration of perchloric acid. The
slope of this correlation gives kH ) 1.5 × 10-3 M-1 s-1 for acid-catalyzed
cleavage of H-1-OH to give H-1+ (eq 2).
We report here the generation of the simple p-quinone me-
thide 1 in dilute aqueous acid, its efficient nucleophilic trapping
by the thiol group of 2-mercaptoethanol, and an analysis of the
resulting kinetic data, which provides the second-order rate
constant for acid-catalyzed cleavage of 4-hydroxybenzyl alcohol
H-1-OH to give the protonated p-quinone methide H-1+. The
data can be combined with kinetic data for the trapping of H-1+
by solvent water24 to obtain values of the Lewis acidity constants
pKR for both H-1+ and 1, along with the overall equilibrium
constant for the 1,6-addition of water to the “extended” carbonyl
group at 1. A comparison with the corresponding data for the
1,2-addition of water to the simple carbonyl group at formal-
dehyde provides insight into the effect of formation of the
aromatic ring in the product on the thermodynamic driving force
for 1,6-addition of water to the p-quinone methide 1.
diode array detector. The detection of HPLC peaks was at 273 nm,
which is λmax for H-1-OH. 3-(4-Methoxyphenyl)-1-propanol (3 × 10-4
M) was used as an internal standard to correct the observed peak areas
for small variations in the injection volume. A ratio of extinction
coefficients ꢀH-1-OH/ꢀH-1-SR ) 1 for H-1-OH and H-1-SCH2CH2OH
at 273 nm was determined by showing that the normalized peak area
for H-1-OH at zero time is identical, within the experimental error of
(5%, with the normalized peak area for the product H-1-SCH2CH2OH
after g10 reaction halftimes. The fraction of H-1-SCH2CH2OH formed
at time t during the course of the reaction, fH-1-SR, was calculated as
the ratio of the normalized peak areas for H-1-SCH2CH2OH at time
t and for H-1-OH at zero time.
Kinetic Studies. Kinetic studies were carried out in 0.1-1.0 M
aqueous perchloric acid in the presence of 0.01-0.15 M 2-mercapto-
ethanol at 25 °C and I ) 1.0 (NaClO4). The concentration of thiol was
determined directly before each kinetic run using Ellman’s reagent [5,5′-
dithiobis-(2-nitrobenzoic acid)].28 The reactions were initiated by
making a 100-fold dilution of a 0.03 M (in the case of faster reactions
with τ1/2 e 25 min) or a 0.3 M (in the case of slower reactions with
τ1/2 > 25 min) solution of H-1-OH in acetonitrile into the reaction
mixture. The reactions were monitored by withdrawal of an aliquot
(100 µL) of the reaction mixture at various times, which was neutralized
with 2 M sodium acetate before analysis by HPLC.
The observed first-order rate constants kobsd (s-1) for reactions with
τ1/2 e 25 min were determined from the disappearance of H-1-OH
and the formation of H-1-SCH2CH2OH as the slopes of semilogarith-
mic plots of reaction progress against time which covered at least 2.5
halftimes. The observed first-order rate constants kobsd (s-1) for reactions
with τ1/2 > 25 min were determined by the method of initial rates as
the slopes of linear plots of the fraction of H-1-OH converted to H-1-
SCH2CH2OH against time, fH-1-SR, that covered no more than 5% of
the total reaction (fH-1-SR e 0.05).
Experimental Section
4-Hydroxybenzyl alcohol (H-1-OH) was reagent grade from Aldrich
and used without further purification. 2-Mercaptoethanol was Bio-
chemica MicroSelect grade from Fluka. All other organic and inorganic
chemicals were reagent grade from commercial sources and used
without further purification. HPLC-grade methanol was used for all
HPLC analyses. Water for kinetic studies and HPLC analyses was
distilled and passed through a Milli-Q water purification system.
Product Analysis by 1H NMR. Product characterization was carried
out on a reaction mixture resulting from reaction of H-1-OH (7 mM)
in the presence of 0.2 M perchloric acid and 0.01 M 2-mercaptoethanol
in 60/40 (v/v) D2O/acetonitrile-d3. After one halftime (4 h), analysis
1
by H NMR spectroscopy at 500 MHz revealed a 1:1 mixture of the
starting material and the thioether substitution product H-1-SCH2CH2OH.
Chemical shifts for H-1-SCH2CH2OH are reported relative to HOD
at 4.67 ppm: δ 7.31, 6.92 (A2B2, 4H, J ) 8.5 Hz, C6H4); 3.82 (2H, s,
ArCH2); 3.74 (2H, t, J ) 6.5 Hz, OCH2); 2.70 (2H, t, J ) 6.5 Hz,
SCH2).
Results
HPLC Analyses. HPLC analysis was carried out as described in
previous work,26,27a except that peak detection was by a Waters 996
In dilute aqueous acid in the presence of 2-mercaptoethanol,
4-hydroxybenzyl alcohol H-1-OH undergoes clean nucleophilic
substitution to give the thioether H-1-SCH2CH2OH, which was
(21) Diao, L.; Yang, C.; Wan, P. J. Am. Chem. Soc. 1995, 117, 5369-5370.
(22) Chiang, Y.; Kresge, A. J.; Zhu, Y. J. Am. Chem. Soc. 2001, 123, 8089-
8094.
1
characterized by H NMR spectroscopy.
(23) Chiang, Y.; Kresge, A. J.; Zhu, Y. J. Am. Chem. Soc. 2002, 124, 717-
Figure 1A shows the dependence of the observed first-order
rate constant kobsd (s-1) for the conversion of H-1-OH to H-1-
SCH2CH2OH in the presence of various concentrations of
722.
(24) Chiang, Y.; Kresge, A. J.; Zhu, Y. J. Am. Chem. Soc. 2002, 124, 6349-
6356.
(25) Richard, J. P. J. Am. Chem. Soc. 1991, 113, 4588-4595.
9
J. AM. CHEM. SOC. VOL. 125, NO. 29, 2003 8815

A R T I C L E S
Substituent Effects on Carbocation Stability
Scheme 2
Scheme 3
Table 1. Parameters Derived from Kinetic Analysis of the
Acid-Catalyzed Conversion of H-1-OH to Give H-1-SCH2CH2OH in
Water (Scheme 2)a
+
b
[H ]/M
k
lim (s-1 c
)
k )
RSH/ks (M-1 d
determining acid-catalyzed cleavage of H-1-OH, klim ) kH[H+],
at high [RSH] (kRSH[RSH] . ks).
0.094
0.445
0.800
1.000
1.34 × 10-4
7.13 × 10-4
1.22 × 10-3
1.48 × 10-3
22
19
24
29
The appearance of the plots in Figure 1A is determined by
two parameters. (1) The limiting first-order rate constant klim
) kH[H+] for reaction at high [RSH], where the acid-catalyzed
cleavage of H-1-OH to give the 4-hydroxybenzyl carbocation
intermediate H-1+ (protonated p-quinone methide) is rate-
determining for the overall nucleophilic substitution. (2) The
rate constant ratio kRSH/ks (M-1) for partitioning of the carboca-
tion H-1+ (protonated p-quinone methide) between nucleophilic
addition of thiol and solvent water (Scheme 2). The values of
kH ) 1.5 × 10-3 M-1 s-1 e
a In the presence of 0.01-0.15 M 2-mercaptoethanol at 25 °C and I )
1.0 (NaClO4). b Concentration of perchloric acid. c Limiting first-order rate
constant for acid-catalyzed reaction of H-1-OH in the presence of relatively
high concentrations of 2-mercaptoethanol (eq 1). d Observed rate constant
ratio for partitioning of the reaction intermediate between nucleophilic
addition of 2-mercaptoethanol and solvent water. e Second-order rate
constant for acid-catalyzed cleavage of H-1-OH to give H-1+, determined
as the slope of the plot of klim against [H+] shown in Figure 1B (eq 2).
k
RSH/ks (M-1) determined for partitioning of H-1+ between
nucleophilic addition of 2-mercaptoethanol and solvent water
range from 22 M-1 at [H+] ) 0.094 M, to 29 M-1 at [H+] )
1.0 M (Table 1). The reaction of solvent water with the
p-quinone methide 1 at these acidities is acid catalyzed and
occurs by nucleophilic attack of water on the protonated quinone
methide H-1+.24 Therefore, the observation that the values of
perchloric acid on the concentration of 2-mercaptoethanol (RSH)
at 25 °C and I ) 1.0 (NaClO4). The least-squares fit of the data
to eq 1 (solid lines), derived for the mechanism shown in
Scheme 2, provided the values of the following parameters that
are reported in Table 1:
k
RSH/ks (M-1) are nearly independent of [H+] shows that the
klim[RSH]
kobsd
)
(1)
(2)
reaction of thiol (kRSH, M-1 s-1) occurs by the same mechanism
(Scheme 2). The small increase in kRSH/ks (M-1) observed at
high [H+] may reflect a medium effect on the reactivity of
solvent water and/or thiol.
(
)
[RSH] + ks/kRSH
klim ) kH[H+]
(1) Values of the limiting first-order rate constant klim (s-1
)
The partitioning ratio kRSH/ks ≈ 20 M-1 can be combined
with the value of ks ) 5.8 × 106 s-1 for the nucleophilic addition
of solvent water to H-1+ 24 to give kRSH ≈ 1.2 × 108 M-1 s-1
as the second-order rate constant for nucleophilic addition of
2-mercaptoethanol to H-1+. A ca. 10-fold larger value of kRSH
) 1.5 × 109 M-1 s-1 has been reported for nucleophilic addition
of 1-propanethiol to the 1-(4-methoxyphenyl)ethyl carbocation
in 50/50 (v/v) trifluoroethanol/water.27b Similarly, the value of
ks ) 5 × 107 s-1 for addition of a solvent of 50/50 (v/v) tri-
fluoroethanol/water to the 1-(4-methoxyphenyl)ethyl carboca-
tion27a is 9-fold larger than ks ) 5.8 × 106 s-1 for addition of
solvent water to H-1+.24
for acid-catalyzed reaction of H-1-OH in the presence of
relatively high concentrations of thiol. Figure 1B shows the
dependence of the values of klim (s-1) on the concentration of
perchloric acid. The slope of this correlation gives the second-
order rate constant kH ) 1.5 × 10-3 M-1 s-1 for acid-catalyzed
cleavage of H-1-OH to give H-1+ (eq 2, derived for Scheme
2).
(2) Values of the rate constant ratio kRSH/ks (M-1) for
partitioning of the 4-hydroxybenzyl carbocation intermediate
H-1+ between nucleophilic addition of 2-mercaptoethanol and
solvent water.
Rate and Equilibrium Constants for Formation and
Reaction of p-Quinone Methide. The second-order rate
constant kH (M-1 s-1) for the specific-acid-catalyzed cleavage
of H-1-OH to give the protonated p-quinone methide H-1+
completes the thermodynamic cycles for the reactions of
p-quinone methide 1 shown in Scheme 3. This rate constant
can be combined with other rate and equilibrium constants from
Scheme 3 to give the following:
Discussion
Reaction Mechanism. In acidic aqueous solution in the
presence of 2-mercaptoethanol (RSH), 4-hydroxybenzyl alcohol
H-1-OH undergoes clean acid-catalyzed conversion to the
thioether H-1-SCH2CH2OH. The data in Figure 1A show that
there is a change in the kinetic order of the reaction with respect
to the concentration of the thiol RSH as the concentration of
RSH is increased. The reaction is first-order in [RSH] when
the concentration of RSH is low but changes to zero-order at
high [RSH]. This shows that the specific-acid-catalyzed nu-
cleophilic substitution reaction of H-1-OH follows the two-
step DN + AN mechanism29 shown in Scheme 2, with a change
from rate-determining addition of thiol to the carbocation
intermediate H-1+ at low [RSH] (ks . kRSH[RSH]) to rate-
(26) (a) Amyes, T. L.; Richard, J. P. J. Am. Chem. Soc. 1990, 112, 9507-
9512. (b) Richard, J. P. J. Am. Chem. Soc. 1989, 111, 1455-1465.
(27) (a) Richard, J. P.; Rothenberg, M. E.; Jencks, W. P. J. Am. Chem. Soc.
1984, 106, 1361-1372. (b) Calculated from the rate constant ratio kaz/
kRSH ) 3.3 for addition of azide ion and 1-propanethiol and kaz ) 5 × 109
M-1 s-1 for the diffusion-limited reaction of azide ion with unstable
carbocations (ref 27a).
(28) Ellman, G. L. Arch. Biochem. Biophys. 1959, 82, 70-77.
(29) Guthrie, R. D.; Jencks, W. P. Acc. Chem. Res. 1989, 22, 343-349.
9
8816 J. AM. CHEM. SOC. VOL. 125, NO. 29, 2003

Richard et al.
A R T I C L E S
(1) A value of pKRH-1 ) -log(ks/kH) ) -9.6 ( 0.130 for the
these benzylic carbocations. We attribute the greater Lewis
Lewis acidity constant of H-1+ can be calculated from kH
)
acidity of Me-1+ compared to H-1+ to the formation of a
1.5 × 10-3 M-1 s-1 for acid-catalyzed cleavage of H-1-OH
(this work) and ks ) 5.8 × 106 s-1 for the reverse nucleophilic
addition of solvent water to H-1+.24
relatively strong hydrogen bond between solvent water and the
acidic phenolic proton of H-1+ (pKa
) -2.0),24 in which
H-1
there is partial proton transfer from the hydrogen bond donor
H-1+ to the hydrogen bond acceptor water.35 This increases both
the formal negative charge density at the phenolic oxygen of
H-1+, relative to that at the methoxy oxygen of Me-1+, and
the stabilization of the cationic center through the formal
delocalization of this charge (Chart 1).
Comparison of 1,2- and 1,6-Addition. The p-quinone
methide 1 is in one sense an extended carbonyl group at which
a cyclohexadiene/phenyl ring has been “inserted” between the
carbonyl oxygen and the methylene group of formaldehyde. The
localization of opposite charges at the 1- and 6-positions of 1
(see Chart 1) is favored relative to this localization of charge at
the 1- and 2-positions of formaldehyde by the compensating
gain in aromaticity of the phenyl ring (Chart 1). Similarly, the
1,6-addition of HNu to 1 is favored relative to 1,2-addition to
formaldehyde by the greater aromaticity of the product p-
(hydroxymethyl)phenol H-1-OH than of the reactant 1 (Scheme
4).
(2) A value of pKadd ) -7.6 ( 0.130 for the overall 1,6-
1
addition of the elements of water to 1 to give H-1-OH can be
H-1
H-1
calculated from the values of pKR
) -9.6 and pKa
)
-2.0 for deprotonation of the phenolic oxygen of H-1+,24
according to eq 3.
1
(3) A value of pKR ) 2.3 ( 0.130 for the Lewis acidity
1
constant of 1 can be calculated from the values of pKadd
-7.6 and pKaP ) 9.931a for deprotonation of the phenolic oxygen
of H-1-OH, according to eq 4.
)
pKadd1 ) pKRH-1 - pKaH-1
pKR1 ) pKadd1 + pKaP
(3)
(4)
Chart 1
Scheme 4
The net effect of this cyclohexadiene/phenyl ring insertion
at the carbonyl group is an increase in the overall equilibrium
constant for the addition of solvent water, from KaddF ) 2.3 ×
These data show that the neutralization of partial negative
charge at the quinone oxygen of 1 by O-protonation to give
H-1+ destabilizes the cationic center by 16 kcal/mol (Chart 1).
By comparison, a value of pKR ) -12.4 has been determined
as the Lewis acidity constant of the 4-methoxybenzyl carboca-
tion Me-1+,33 so that O-methylation of 1 to give Me-1+ results
in an even larger 20 kcal/mol destabilization of the electrophilic
center toward addition of solvent water (Chart 1).
The effect of a methyl for hydrogen substitution on the
Brønsted acidity of nitrogen or oxygen acids is generally small
for acid-base reactions in water. For example, the pKas of water
and methanol are similar, and only small changes in the acidity
of the ammonium ion are observed as three of the four
hydrogens are replaced by methyl groups.34 By comparison, the
methyl for hydrogen substitution at H-1+ to give Me-1+ occurs
at a site distant from the Lewis-acid-type addition of water to
103 for hydration of formaldehyde,36 to Kadd ) 4.0 × 107 for
1
hydration of the p-quinone methide 1 (this work, Table 2), so
that KT ) Kadd1/Kadd ) 1.7 × 104 for transfer of the elements
F
of water from formaldehyde hydrate to 1 (Scheme 4). The
relatively small driving force of 6 kcal/mol for this transfer of
water from CH2(OH)2 to 1 represents the balance between much
larger opposing effects:
(1) The ca. 31 kcal greater stability of the reactant formal-
dehyde hydrate than of the product p-(hydroxymethyl)phenol
H-1-OH due to the stronger single bonds to oxygen. This is a
result of the 15 kcal/mol stabilizing interactions between the
geminal oxygens at CH2(OH)2,37 and the weakening of the
phenolic O-H bond at H-1-OH due to the ca. 16 kcal/mol
stabilization of the alkoxy radical by the aromatic ring.38
(2) The even larger ca. (31 + 6) ) 37 kcal/mol driving force
associated with the larger formal aromatic stabilization of the
six-membered ring at the product H-1-OH than at the reactant
1. It is interesting that this effect is very similar to the total
aromatic stabilization of a phenyl ring, which has been estimated
to lie between 30 and 36 kcal/mol.39 It might be argued that
any formal contribution of an aromatic zwitterionic valence
(30) Quoted errors are standard errors that were calculated as described in
Supporting Information.
(31) (a) Calculated from pKa ) 10.0 for phenol (ref 34), σ ) 0.03 for the p-CH2-
OMe substituent (ref 32, p 66), and F ) 2.2 for ionization of ring-substituted
phenols (ref 32, p 162). (b) Calculated from pKa ) -3.0 for protonated
4-methoxybenzyl alcohol, the difference in the values of σn ) -0.5 and
-0.13 for the 4-O- and 4-MeO substituents (ref 32, p 72), respectively,
and F ) 1.1 for ionization of ring-substituted acetophenone hydrates
(Stewart, R.; Linden, R. V. D. Can. J. Chem. 1960, 38, 399-406) or ring-
substituted benzylammonium ions (Blackwell, L. F.; Fischer, A.; Miller, I.
J.; Topsom, R. D.; Vaughan, J. J. Chem. Soc. 1964, 3588-3591). The value
of pKa ) -3.0 for protonated 4-methoxybenzyl alcohol was calculated
from pKa ) -2.05 for MeOH2+ (Perdoncin, G.; Scorrano, G. J. Am. Chem.
Soc. 1977, 99, 6983-6986), σI ) 0.11 for the 4-MeOC6H4 substituent
(Charton, M. Prog. Phys. Org. Chem. 1981, 13, 119-251), and FI ) 8.75
(footnote 9 of ref 42) for ionization of alcohols of structure R1R2CHOH.
(c) Calculated as described for the pKa for protonated 4-methoxybenzyl
alcohol in ref 31b, using the values of σI ) 0.28 and -0.01 for the EtO
and Et substituents, respectively.
(35) Stahl, N.; Jencks, W. P. J. Am. Chem. Soc. 1986, 108, 4196-4205.
(36) (a) Funderburk, L. H.; Aldwin, L.; Jencks, W. P. J. Am. Chem. Soc. 1978,
100, 5444-5459. (b) The literature for the experimentally determined
equilibrium constants for hydration of formaldehyde is summarized in
footnote 25 of ref 36a.
(37) Benson, S. W. Angew. Chem., Int. Ed. Engl. 1978, 17, 812-819.
(38) The bond dissociation energy D(RO-H) for the O-H bond of methanol
is 16 kcal/mol larger than that for the O-H bond of phenol [calculated
from D(RO-H) ) ∆fH°(H•) + ∆fH°(RO•) - ∆fH°(ROH); standard heats
of formation ∆fH° are from NIST Standard Reference Database Number
69, July 2001 Release (http://webbook.nist.gov/chemistry/)].
(39) Schleyer, P. v. R.; Manoharan, M.; Jiao, H.; Stahl, F. Org. Lett. 2001, 3,
3643-3646.
(32) Hine, J. Structural Effects on Equilibria in Organic Chemistry; Wiley: New
York, 1975.
(33) Toteva, M. M.; Richard, J. P. J. Am. Chem. Soc. 2002, 124, 9798-9805.
(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.
9
J. AM. CHEM. SOC. VOL. 125, NO. 29, 2003 8817

A R T I C L E S
Substituent Effects on Carbocation Stability
Table 2. Rate and Equilibrium Constants and Marcus Intrinsic Barriers for Addition of Solvent Water to 1 and Formaldehyde and to Their
O-Alkylated Analogues Me-1+ and 2 (Scheme 5)a
c
d
e
f
electrophile
ks/s-1 b
K
add
K
int
K
Z
∆ log K
z
Λg
1
3.3h
4.0 × 107
3.2 × 1012 i
1.0 × 103 Mk
4.0 × 1012 m
2.5 × 104 Mp
1.3 × 10-5
2.3 × 109
5.8 × 10-10
2.2 × 105
13.2 ( 0.8
11.5 ( 0.6
8.7 ( 1.2
6.6 ( 0.8
14.2
14.6
Me-1+
H2CO
2
2.5 × 108 j
2.3 × 1012 Mj
2.3 × 103 l
10l
2 × 1010 n
5.4 × 109 Mo
a In water at 25 °C. b First-order rate constant for addition of solvent water to the electrophile. c Overall equilibrium constant for addition of solvent water
to give the neutral alcohol product. These are dimensionless quantities for 1 and H2CO but have units of M for Me-1+ and 2 for which the addition of water
generates a proton. d Equilibrium constant for the proton-transfer reaction that converts the initial ionic solvent adduct to the final neutral alcohol product.
These are dimensionless quantities for 1 and H2CO but have units of M for Me-1+ and 2. e Equilibrium constant for addition of solvent water to the
electrophile to give the ionic solvent adduct, calculated using the relationship KZ ) Kadd/Kint.
f Difference in log Kz for Me-1+ and 1 or 2 and H2CO.
g Marcus intrinsic barrier for addition of solvent water to the electrophile, calculated from the values of ks and KZ using eq 5. Quoted errors are standard
errors that were calculated as described in Supporting Information. h Data from ref 24. i Calculated from the values of K1 ) 102.6 for deprotonation of
-OC6H4CH2OH2+ [ref 31b] and K2 ) 10-9.9 for deprotonation of the phenolic oxygen of HOC6H4CH2OH [ref 31a] using the relationship KZ ) K1/K2.j Data
from ref 33. k From the estimated value of pKa ) -3.0 for protonated 4-methoxybenzyl alcohol (ref 31b). l Data from ref 36a. m Calculated from the values
+
of K1 ) 10-0.7 for deprotonation of -OCH2OH2 (ref 36a) and K2 ) 10-13.3 for deprotonation of HOCH2OH (ref 36a) using the relationship KZ ) K1/K2.
n Data from ref 46. o Calculated using the relationship Kadd ) (ks/kH) with ks ) 2 × 1010 s-1 (ref 46) and the estimated value of kH ) 3.7 M-1 s-1 for the
acid-catalyzed cleavage of propionaldehyde ethyl hemiacetal to give the oxocarbenium ion 2. This value of kH was estimated from kH ) 1.63 M-1 s-1 for
the acid-catalyzed hydrolysis of acetaldehyde diethyl acetal (ref 47) and the 2.3-fold faster acid-catalyzed cleavage of 1-(4-methoxyphenyl)ethyl alcohol than
of the corresponding ethyl ether in 50/50 (v/v) trifluoroethanol/water (footnote 22 of ref 14). p From the estimated value of pKa ) -4.4 for protonated
propionaldehyde ethyl hemiacetal (ref 31c).
Scheme 5
bond resonance form to the structure of 1 (Chart 1) should
reduce the driving force for formation of H-1-OH associated
with the formation of an aromatic ring. We therefore suggest
that the stabilization of 1 toward the addition of water by the
development of “aromatic” character is roughly offset by a
compensating destabilization of 1 due to weakening of the Cd
O π-bond [there is no such bond at the formal zwitterion]
relative to that at formaldehyde. Thus, the apparent effect of
aromatization on the thermodynamic driving force for the
isodesmic reaction shown in Scheme 4 is approximately equal
to the full aromatic stabilization of the product H-1-OH.
Intrinsic Reaction Barriers. The kinetic barriers to organic
reactions depend on both the thermodynamic driving force to
the reaction and the “intrinsic” barrier to the reaction in the
absence of any driving force (∆Go ) 0).40,41 Thus, the
determination of both the thermodynamic driving force ∆Go
and the Marcus intrinsic barrier Λ is critical to the development
of explanations for why different reactions proceed at different
rates. There have been relatively few quantitative determinations
of the intrinsic barriers for organic reactions.12,42 We are
interested in characterizing these barriers for a large body of
simple organic reactions, to develop an understanding of their
origin through the application of simple chemical principles and,
perhaps, through computational studies by others aimed at
reproducing the experimental intrinsic reaction barriers.43
Table 2 summarizes the rate and equilibrium constants and
estimated Marcus intrinsic barriers Λ for the nucleophilic
addition of solvent water to the series of simple electrophiles
shown in Scheme 5. These values of Λ were determined from
the rate constants ks (s-1) for the addition of solvent water and
the equilibrium constants KZ for formation of the initial ionic
solvent adduct (Scheme 5), using the Marcus equation derived
at 298 K (eq 5). The values of KZ were calculated from Kadd
for the overall addition of solvent water to the electrophile to
form the neutral alcohol adduct and Kint for conversion of the
initial ionic solvent adduct to the neutral alcohol product
(estimated as described in the footnotes to Table 2), using the
relationship KZ ) Kadd/Kint (Scheme 5).
2
1.36log KZ
4Λ
1
1.36
log ks )
17.44 - Λ 1 -
(5)
{
(
) }
O-Methylation of 1 to give Me-1+ results in a 19 kcal/mol
increase in the driving force for addition of solvent water to
give the initial ionic solvent adduct (∆log KZ ) 14.2) and a
small 1.7 kcal/mol decrease in the Marcus intrinsic barrier Λ
to the reaction (Table 2). This is consistent with the notion that
the intrinsic barrier to carbocation-nucleophile combination is
sensitive to changes in the thermodynamic driving force
resulting from changes in electron donation from substituents
that are distant from the cationic center. There will be a
“weaker” donation of electrons from the p-oxygen at Me-1+
than at 1, because electron donation from oxygen at Me-1+ has
the effect of increasing the formal positive charge density at
the electronegative oxygen. The combination of O- and R-ethyl-
ation of formaldehyde to give the simple oxocarbenium ion 2
(40) Marcus, R. A. J. Phys. Chem. 1968, 72, 891-899.
(41) Marcus, R. A. J. Am. Chem. Soc. 1969, 91, 7224-7225.
(42) Richard, J. P.; Williams, K. B.; Amyes, T. L. J. Am. Chem. Soc. 1999,
121, 8403-8404.
(43) Schindele, C.; Houk, K. N.; Mayr, H. J. Am. Chem. Soc. 2002, 124, 11208-
11214.
9
8818 J. AM. CHEM. SOC. VOL. 125, NO. 29, 2003

Richard et al.
A R T I C L E S
results in a 20 kcal/mol increase in the driving force for addition
of solvent water (∆log KZ ) 14.6) and a small 2.1 kcal/mol
decrease in the Marcus intrinsic barrier Λ to the reaction (Table
2).44
between the incoming nucleophile water and the electrophilic
carbon and the development of unfavorable (destabilizing)
interactions that accompanies this partial bond formation. The
latter include the development of positive charge at the oxygen
of the water nucleophile, partial cleavage of the CdO double
bond of the carbonyl group at formaldehyde (see 3), and the
much more extensive electronic reorganization that occurs at
the extended π-system at 1 (see 4), which is the ultimate cause
of the ca. 5 kcal/mol larger intrinsic barrier for 1,6- than for
1,2-addition of water. The deeper relationship between the
changes in bonding that occur on proceeding from reactant to
the transition state for these two reactions might be better
understood through theoretical and/or computational studies
designed to model these intrinsic barriers.
Despite the 6 kcal/mol greater thermodynamic driving force
for the addition of solvent water to the extended carbonyl group
at 1 than to the simple carbonyl compound formaldehyde, these
reactions have almost identical rate constants of ks ) 3.3 and
10 s-1, respectively (Table 2). This corresponds to a 4.5 kcal/
mol larger Marcus intrinsic barrier for addition of solvent water
to 1 (Λ ) 13.2 kcal/mol) than to formaldehyde (Λ ) 8.7 kcal/
mol), which shows that 1,6-addition to a p-quinone methide is
intrinsically 2000-fold slower than 1,2-addition to a carbonyl
group (Table 2). There is a very similar difference of 4.9 kcal/
mol in the intrinsic barriers for addition of solvent water to Me-
1+ (Λ ) 11.5 kcal/mol) and the simple oxocarbenium ion 2 (Λ
) 6.6 kcal/mol). We conclude that alkylation of the carbonyl
groups at 1 and formaldehdyde to give Me-1+ and a simple
oxocarbenium ion, respectively, results in very little change in
the relative Marcus intrinsic barriers for the addition of solvent
water to these electrophiles.44 This provides strong evidence
that O-alkylation, a dramatic structural modification, has only
a small effect on the steepness of the curvature of the reaction
coordinate profile for nucleophile addition, which is the primary
determinant of the magnitude of Λ for simple organic reac-
tions.12,45
Acknowledgment. We acknowledge National Institutes of
Health Grant GM 39754 for generous support of this work. The
manuscript was prepared during the term of J.P.R. as a SFI
Walton Fellow at University College, Dublin, Ireland. This paper
is dedicated to the memory of Mike Moran, who passed away
suddenly during the course of this work.
Supporting Information Available: Discussion of the un-
1
H-1
certainties and standard errors in the values of pKR , pKR
,
and pKadd1 (Scheme 3) and in the rate and equilibrium constants
and Marcus intrinsic barriers reported in Table 2. This material
is available free of charge via the Internet at http://pubs.acs.org.
JA029588V
(44) O-methylation of formaldehyde gives the highly unstable methoxymethyl
carbocation which undergoes addition of solvent water with ks g 5 × 1012
s-1 (ref 46) so that there is little or no chemical barrier to this reaction.
Therefore, to assess the effect of O-alkylation of formaldehyde on the
intrinsic barrier for the addition of water, we use 2 as a prototypical simple
oxocarbenium ion. We have estimated values of Kz ) 2.4 × 106 M and Λ
) 10.4 kcal/mol for addition of solvent water to the 1-(4-methoxyphenyl)-
ethyl carbocation to give the initial ionic solvent adduct, using the
methodology described in this paper. Comparison of these quantities with
those for Me-1+ (Table 2) shows that the presence of the R-alkyl group at
2 is expected to result in a decrease in the thermodynamic driving force
for the addition of solvent water of ca. 4 kcal/mol, but that it should have
very little effect on the Marcus intrinsic barrier for this reaction.
By comparison, the 4.5 kcal/mol difference in the intrinsic
barriers for addition of solvent water to 1 and formaldehyde is
almost as large as the 6 kcal/mol difference in the thermody-
namic driving force for these reactions (Table 2). The intrinsic
barriers for the addition of water to formaldehyde and to 1
represent the balance in the transition state between the
stabilization that results from the formation of a partial bond
(45) Richard, J. P. Tetrahedron 1995, 51, 1535-1573.
(46) Amyes, T. L.; Jencks, W. P. J. Am. Chem. Soc. 1989, 111, 7888-7900.
(47) Kresge, A. J.; Weeks, D. P. J. Am. Chem. Soc. 1984, 106, 7140-7143.
9
J. AM. CHEM. SOC. VOL. 125, NO. 29, 2003 8819