Kinetics of proton transfer from 2-nitro-4-X-phenylacetonitriles to piperidine and morpholine in aqueous Me2SO. Solvent and substituent effects on intrinsic rate constants. Transition state imbalances

Article abstract of DOI:10.1021/ja961837q

Rate constants (k₁(B)) for the deprotonation of 2-nitro-4-X-phenylacetonitrile, 2-X (X = NO₂, SO₂CH₃, CN, CF₃, Br, and Cl) by piperidine and morpholine and for the reverse reaction (k_-1(BH)) have been determined in 90% Me₂SO- 10% water, 50% Me₂SO-50% water, and water (X = NO₂, SO₂CH₃, CN only). Bronsted β(B) values (dlog k₁(B)/dpK(a)(BH)), Bronsted α(CH) values (dlog k₁(B)/dlog K(a)(CH)), and intrinsic rate constants (log k(o) = log(k₁/q) for pK(a)(BH)-p K(a)(CH) + log(p/q) = 0) were calculated from these data. α(CH) is smaller than β(B), implying an imbalance wnich is consistent with a transition state in which delocalization of the negative charge into the 2-nitrophenyl moiety lags behind proton transfer. A consequence of this imbalance is that the intrinsic rate constant decreases with increasing electron withdrawing strength of X. For π-acceptor substituents (NO₂, SO₂CH₃, CN) there is a further decrease in k(o) due to a lag in the delocalization of the charge into X. The intrinsic rate constants depend very little on the Me₂SO content of the solvent which is shown to be the result of compensation of mainly two competing factors. One is the stabilization of the polarizable transition state by the polarizable Me₂SO which increases k(o); the other is attributed to a lag in the solvation of the developing carbanion behind proton transfer at the transition state which leads to a decrease in k(o).

Full text of DOI:10.1021/ja961837q

11446
J. Am. Chem. Soc. 1996, 118, 11446-11453
Kinetics of Proton Transfer from
2-Nitro-4-X-phenylacetonitriles to Piperidine and Morpholine in
Aqueous Me2SO. Solvent and Substituent Effects on Intrinsic
Rate Constants. Transition State Imbalances
Claude F. Bernasconi* and Philip J. Wenzel
Contribution from the Department of Chemistry and Biochemistry of the UniVersity of California,
Santa Cruz, California 95064
ReceiVed May 31, 1996X
Abstract: Rate constants (kB) for the deprotonation of 2-nitro-4-X-phenylacetonitrile, 2-X (X ) NO2, SO2CH3, CN,
1
CF3, Br, and Cl) by piperidine and morpholine and for the reverse reaction (kB-H1) have been determined in 90%
Me2SO-10% water, 50% Me2SO-50% water, and water (X ) NO2, SO2CH3, CN only). Brønsted âB values (dlog
kB1 /dpKaBH), Brønsted RCH values (dlog kB1 /dlog KaCH), and intrinsic rate constants (log ko ) log(k1/q) for pKBH - p
a
KCa H + log(p/q) ) 0) were calculated from these data. RCH is smaller than âB, implying an imbalance which is
consistent with a transition state in which delocalization of the negative charge into the 2-nitrophenyl moiety lags
behind proton transfer. A consequence of this imbalance is that the intrinsic rate constant decreases with increasing
electron withdrawing strength of X. For π-acceptor substituents (NO2, SO2CH3, CN) there is a further decrease in
ko due to a lag in the delocalization of the charge into X. The intrinsic rate constants depend very little on the
Me2SO content of the solvent which is shown to be the result of compensation of mainly two competing factors.
One is the stabilization of the polarizable transition state by the polarizable Me2SO which increases ko; the other is
attributed to a lag in the solvation of the developing carbanion behind proton transfer at the transition state which
leads to a decrease in ko.
When dealing with chemical reactivity it is good practice to
separate thermodynamic driving force from a purely kinetic
factor commonly referred to as the intrinsic barrier (∆Gq0 ) ∆
Gq1 ) ∆Gq-1 when ∆G° ) 0) or the intrinsic rate constant (ko )
k1 ) k-1 when K1 ) 0).1 Indeed, many important insights
regarding the factors that influence proton transfer from carbon
acids to a variety of bases have been gained from the
determination of the intrinsic rate constants of these reactions.2
A major conclusion to be drawn from a recent compilation of
some 40 examples2b is that ko progressively decreases with
increasing resonance stabilization of the resulting carbanion.
This decrease in ko has been attributed to a lag in charge
delocalization behind proton transfer at the transition state.
Another important factor that can substantially reduce ko is a
lack of synchronization, at the transition state, between the
proton transfer and the solvation of the products or desolvation
of the reactants, in particular the lag in the solvation of the
developing carbanion behind the proton transfer.2a,b
by the piperidine/morpholine pair4,5 and carboxylate ions,5
respectively, show the very large ko-depressing effect of non-
synchronous resonance development and solvation. That the
lag in the solVation of the developing nitronate ion contributes
substantially to the overall lowering of ko can be seen from the
much higher log ko values in 90% or 100% Me2SO (log ko )
1.75 and 1.88 for deprotonation by R2NH and RCOO-,
respectively, in 90% Me2SO-10% water;5 log ko ) 2.81 for
deprotonation by ArCOO- in pure Me2SO6) which are consistent
with the much weaker solvation of the nitronate ion by Me2SO
than by water. Similar results have been reported for the
deprotonation of nitromethane.5
A dramatic illustration of these points comes from the
comparison of phenylnitromethane with 1. For 1, log ko ∼ 9.0
for deprotonation by carboxylate ions in water3 represents one
extreme which suggests that 1 behaves essentially like a normal
acid with no charge delocalization or resonance stabilization
of its conjugate base, and solvation effects that are too small to
significantly affect ko. At the other extreme, log ko ) -1.22
and -2.10 for the deprotonation of phenylnitromethane in water
Acetylacetone takes an intermediate position between 1 and
phenylnitromethane, with log ko ) 3.14 for deprotonation by
RCOO- and log ko ) 2.60 for deprotonation by piperidine/
morpholine in aqueous solution, and log ko ) 5.27 (RCOO-)
and 3.64 (pip/mor) in 90% Me2SO-10% water.8 The inter-
mediate values for log ko are consistent with a smaller degree
of resonance stabilization of the acetylacetonate ion compared
to the phenylnitromethane anion; and the smaller solvent effect
on log ko is consistent with the smaller solvational stabilization
X Abstract published in AdVance ACS Abstracts, November 1, 1996.
(1) For a reaction with a forward rate constant k1, a reverse rate constant
k-1 and an equilibrium constant K1.
(2) For recent reviews, see (a) Bernasconi, C. F. Acc. Chem. Res. 1987,
20, 301. (b) Bernasconi, C. F. AdV. Phys. Org. Chem. 1992, 27, 119. (c)
Bernasconi, C. F. Acc. Chem. Res. 1992, 25, 9.
(4) Bordwell, F. G.; Boyle, W. J., Jr. J. Am. Chem. Soc. 1972, 94, 3907.
(5) Bernasconi, C. F.; Kliner, D. A. V.; Mullin, A. S.; Ni, J. X. J. Org.
Chem. 1988, 53, 3342.
(6) Based on data in reference 7.
(7) Keeffe, J. R.; Morey, J.; Palmer, C. A.; Lee, J. C. J. Am. Chem. Soc.
1979, 101, 1295.
(3) Washabaugh, M. W.; Jencks, W. P. J. Am. Chem. Soc. 1989, 111,
674, 683.
(8) Bernasconi, C. F.; Bunnell, R. D. Isr. J. Chem. 1985, 26, 420.
S0002-7863(96)01837-9 CCC: $12.00 © 1996 American Chemical Society

Deprotonation of 2-Nitro-4-X-phenylacetonitriles
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11447
by water of the acetylacetonate compared to the phenylnitronate
anion.9 1,3-Indandione shows very similar behavior as acetyl-
acetone.10
meter for X ) NO2, CN, CH3SO2, CF3, Br and Cl and 90%
Me2SO-10% water (v/v) and 50% Me2SO-50% water (v/v),
and for X ) NO2, CN and CH3SO2 in water, at 20 °C. With
X ) CF3, Br and Cl in water the acidity of 2-X was too low
for such measurements. All experiments were conducted under
pseudo-first-order conditions in excess amine buffer.
A qualitatively different situation has been observed with
9-cyanofluorene,11 9-carbomethoxyfluorene11 and (R-cyano-
diphenylmethane)bis(tricarbonylchromium(0)).12 For example,
deprotonation of 9-cyanofluorene by primary amines in 10%
Me2SO-90% water is characterized by log ko ) 3.62 which is
not much larger than for acetylacetone and again places it about
midway between 1 and phenylnitromethane. Here, however,
most of the reduction in ko relative to that for 1 comes from the
resonance effect of the fluorenyl ion while solvation is relatively
unimportant. This can be seen from the fact that a change to
90% Me2SO-10% water leaves ko essentially unaffected. This
finding is consistent with the 9-cyanofluorenyl anion being less
solvated in the water-rich than in the Me2SO-rich solvent,9 which
is a consequence of the negative charge not being concentrated
on oxygen atoms that can be solvated by hydrogen bonding
(nitronate and enolate ions), but being strongly dispersed over
a large π-system.
The pseudo-first-order rate constants for equilibrium approach
is given by eq 2 which includes terms for the water, H+ and
OH- reactions. These latter terms were either negligible
H
kobsd ) kH1 O + k-1aH+ + kOHaOH- + k-H21O + kB1 [B] +
2
1
BH
k
[BH+] (2)
-1
(especially in 90% Me2SO) or resulted in a small intercept when
kobsd was plotted vs [B] or [BH+]; k1B and kBH which are the
-1
focus of this study were determined from the slopes of these
plots as follows.
When pKBa H . pKCa H (CH refers to 2-X), the slope of a plot
of kobsd vs [B] yielded k1B directly and kBH was calculated as
An interesting but as yet unexplored case with respect to
solvent effects on ko is the deprotonation of 2-nitro-4-X-
phenylacetonitrile (2-X). The only member of this series for
which a few relevant kinetic data are available is 2,4-dinitro-
phenylacetonitrile (2-NO2). In 50% Me2SO-50% water log
-1
BH
-1
k
) kB1 KaBH/KCa H. This was the case for the reactions of
piperidine with 2-NO2, 2-SO2CH3, 2-CN and 2-CF3 in 90%
Me2SO, the reaction of morpholine with 2-NO2 and 2-SO2CH3
in 90% Me2SO, the reactions of piperidine with 2-SO2CH3 and
2-CN in 50% Me2SO, and the reaction of piperidine with 2-NO2
in water. When pKBa H , pKaCH the slope of a plot of kobsd vs
[BH+] afforded k-BH1 directly and k1B was obtained as kB1
)
BH
k
KCa H/KBa H
. The reactions of morpholine with 2-Br and
-1
2-Cl in 90% Me2SO, of morpholine with 2-CF3, 2-Br and 2-Cl
in 50% Me2SO, and of 2-SO2CH3 and 2-CN with morpholine
in water fall into this category. When pKCa H ∼ pKaBH the slope
of a plot of kobsd vs [B] is pH dependent and given by kB1 +
ko ) 2.75 with piperidine/morpholine13 places it close to
acetylacetone in terms of the combined effects of resonance and
solvation. In the carbanion, 2-NO2-, there is undoubtedly a
significant amount of charge delocalization over the extended
π-system which makes it similar to the 9-cyanofluorenyl anion.
On the other hand, a good portion of this charge is likely to be
concentrated on the two nitro groups of 2-NO2- which should
allow for hydrogen bonding solvation in a water-rich solvent.
Hence the solvent effect on ko for deprotonation of 2-NO2 may
be somewhere between that for the deprotonation of acetylac-
etone and 9-cyanofluorene. A major objective of this paper is
to test this prediction, to determine the relative solvation energies
BH
k
k
aH /KBa H. A plot of this slope vs aH then yields kB1 and
+
+
-1
/KBa H, and kBH can be obtained from kBH/KBa H and the known
BH
-1
-1
-1
KBa H. This situation prevailed in the reactions of morpholine
with 2-CN and 2-CF3 in 90% Me2SO, the reactions of piperidine
with 2-Cl and 2-Br in 90% Me2SO, the reactions of morpholine
with 2-SO2CH3 and 2-CN and of piperidine with 2-CF3 in 50%
Me2SO, and the reactions of morpholine with 2-NO2 and of
piperidine with 2-SO2CH3 and 2-CN in water. A representative
example is shown in Figure 1S of the supporting information14
where kobsd for the reaction of 2-Cl with piperidine is plotted
versus piperidine concentrations at various pH values. The plot
-
14
(solvent transfer coefficients)9 for 2-NO2 in various Me2SO-
of the slopes for Figure 1S versus aH is shown in Figure 2S.
+
water mixtures, and analyze the solvent effect on ko based on
these solvation energies.
The various rate constants determined as described above are
summarized in Table 1 while the raw data are reported
elsewhere.15
A second objective is to examine how a change in the
X-substituent of 2-X affects the rate of proton transfer, thereby
providing information about (a) possible transition state imbal-
ance, (b) changes in transition state structure, (c) substituent
effects on ko, and (d) substituent dependence of the solvent effect
on ko.
The pKCa H values of the various 2-X in the three solvents
were measured spectrophotometrically. They are included in
Table 1. In those cases where both kB1 and kBH could be
-1
determined from the pH-dependence of the slopes of kobsd vs
[B] the agreement between the kinetically and spectrophoto-
metrically obtained pKCa H values was very good.
Results
Rate and pKa Determinations. Rate constants kB1 and
Transfer Activity Coefficients. Activity coefficients for the
50
transfer of 2-X from water to 50% Me2SO (0γ ), from water
BH
CH
k
for the reaction in eq 1 with B being piperidine and
-1
90
CH
to 90% Me2SO (0γ ), and from 50% to 90% Me2SO (50
morpholine were determined in a stopped-flow spectrophoto-
γ9C0H) were determined from partition experiments with p-
(9) Bernasconi, C. F.; Bunnell, R. D. J. Am. Chem. Soc. 1988, 110, 2900.
(10) Bernasconi, C. F.; Paschalis, P. J. Am. Chem. Soc. 1986, 108, 2969.
(11) Bernasconi, C. F.; Terrier, F. J. Am. Chem. Soc. 1987, 109, 7115.
(12) Bernasconi, C. F.; Bunnell, R. D.; Terrier, F. J. Am. Chem. Soc.
1988, 110, 6514.
xylene, as detailed in the Experimental Section. In conjunction
(14) See paragraph concerning supporting information at the end of this
paper.
(13) Bernasconi, C. F.; Hibdon, S. A. J. Am. Chem. Soc. 1983, 105,
4343.
(15) Wenzel, P. J. Ph.D. Thesis, University of California, Santa Cruz,
1996.

11448 J. Am. Chem. Soc., Vol. 118, No. 46, 1996
Bernasconi and Wenzel
Table 1. Rate Constantsa for Proton Transfer between 2-X and
Amines at 20 °C
Table 2. Transfer Activity Coefficients for the Transfer of 2-X
and Some Other Carbon Acids from Water to 50% and 90% Me2SO
and from 50% to 90% Me2SO at 20 °C
piperidineb
morpholinec
CH
log0γ5C0H
log50γ9C0H
log0γ9C0H
BH
BH
kB1 ,
k
,
kB1 ,
k ,
-1
-1
2-X
pKCa H
M-1 s-1
M-1 s-1
M-1 s-1
M-1 s-1
2-NO2
2-SO2CH3
2-CN
2-CF3
2-Br
-0.70
-0.82
-0.70
-0.59
-0.38
-0.99
-1.26
-0.55
-2.09
-1.95
-1.77
-2.24
<-2.53
-2.24
-1.21
-0.43
-1.75
-0.77
-1.68
-2.08
-1.25
-2.68
-2.33
90% Me2SO-10% Waterd
2-NO2
6.00 ( 0.02 1.85 × 105 3.27 × 100 1.92 × 104 2.36 × 101
2-SO2CH3 7.85 ( 0.02 7.39 × 104 9.52 × 101 7.86 × 103 6.85 × 102
2-CN
2-CF3
2-Br
8.14 ( 0.04 5.48 × 104 1.38 × 102 6.02 × 103 1.02 × 103
9.24 ( 0.02 2.76 × 104 8.76 × 102 2.59 × 103 5.53 × 103
11.03 ( 0.02 6.29 × 103 1.22 × 104 4.21 × 102 5.55 × 104
11.15 ( 0.05 5.52 × 103 1.42 × 104 3.60 × 102 6.26 × 104
2-Cl
9-CN-Fla,b
[(CO)3C6H5]2CHCNc
9-COOMe-Fla,b
1,3-indandioneb
-1.64d
-3.88e
2-Cl
-1.90d
-0.39d
0.21
-1.11
-0.10
-4.14e
-1.60e
-0.22
-2.86
-0.87
50% Me2SO-50% Watere
b
CH2(COCH3)2
PhCH2NO2
CH3NO2
2-NO2
8.06
2.37 × 104 2.43 × 101 1.71 × 103 3.74 × 102
f
b
2-SO2CH3 9.40 ( 0.02 1.10 × 104 2.46 × 102 9.47 × 102 4.54 × 103
b
2-CN
2-CF3
2-Br
9.53 ( 0.03 9.70 × 103 2.93 × 102 8.44 × 102 5.45 × 103
10.71 ( 0.03 4.78 × 103 2.71 × 103 3.20 × 102 3.13 × 104
12.02 ( 0.04 1.54 × 103 1.44 × 104 5.66 × 101 1.13 × 105
12.17 ( 0.02 1.26 × 103 1.66 × 104 4.33 × 101 1.22 × 105
a Fl ) fluorene. b Reference 9. c (R-Cyanodiphenylmethane)bis(tri-
carbonylchromium(0)), ref 12. d log10γ5C0H e log10γC90H
.
.
2-Cl
Watere
Table 3. Transfer Coefficients for the Transfer of 2-X- and Some
Other Carbanions from Water to 50% and 90% Me2SO and from
50% to 90% Me2SO
2-NO2
9.91 ( 0.03 3.94 × 103 9.45 × 101 3.69 × 102 3.78 × 103
11.43 ( 0.02 2.39 × 103 1.90 × 103 1.62 × 102 5.41 × 104
2-SO2CH3 11.12 ( 0.05 2.65 × 103 1.02 × 103 2.31 × 102 3.83 × 104
2-CN
C-
log0γ5C0
log50γ9C0
lo50γ9C0
-
-
-
a Error limits estimated at (4% or better. b pKBa H ) 10.74 (90%
-
2-NO2
-0.62
-0.61
-0.67
-1.93
-1.39
-1.11
-2.44
-1.82
-1.67
-2.65
<-2.56
-2.12
1.38
-2.54
-2.00
-1.78
Me2SO), 11.05 (50% Me2SO), 11.53 (water). c pKaBH ) 8.91 (90%
Me2SO), 8.72 (50% Me2SO), 8.90 (water). d µ ) 0.06 M (KCl). e µ )
0.5 M (KCl). f Reference 13.
-
2-SO2CH3
2-CN-
-
2-CF3
2-Br-
2-Cl-
16
+
with the transfer activity coefficient of the hydronium ion (γH
)
9-CN-Fl- a,b
[(CO)3C6H5]2CCN- c
9-COOMe-Fl- a,b
1,3-indandione- b
CH(COCH3)2
PhCHdNO2
-1.37d
-4.02e
and the solvent dependence of pKCa H, the transfer activity
50
coefficients for 2-X- (0γC-, γ90-, and 50γC90-) were then ob-
0
C
0.41d
2.36
1.99
2.87
1.79e
5.03
4.09
6.70
tained from equations such as eq 3.
- b
2.67
2.10
3.83
- b
log0γ5C0- ) 0∆50pKCH - log0γH50+ + log 0γ50
(3)
a
CH
- b
CH2dNO2
a Fl ) Fluorene. b Reference 9. c Reference 12. d log10γC50 e log10
.
-
The various log γCH values for 2-X and some other carbon
acids are summarized in Table 2, the log γC- values for 2-X-
and other carbanions in Table 3. Note that we have adopted
Parker’s17 definition of transfer coefficients as illustrated in eq
γ9C0
.
-
rene and (R-cyanodiphenylmethane)bis(tricarbonylchromium-
(0)) (Tables 2 and 3). This contrasts with the nitronate and
enolate ions where the negative charge is more localized on
oxygen which leads to stronger solvation by water than by Me2-
90
0
0
4 for γC-, where ∆90Gtr(C-) is
log0γ9C0- ) 0∆90G (C-)/2.303RT
(4)
tr
SO (positive log0γC50-, log50γ90-, etc., Table 3) and hence to an
C
the free energy of transfer of C- from water to 90% Me2SO.
increase in pKCa H in the Me2SO-containing solvents.18
90
-
0
Thus a negative log γC means that C- is more strongly, a
Substituent Effects on Acidities of 2-X. As one would
expect, the acidity of 2-X increases with increasing electron
withdrawing strength of X. This increase is quite substantial
and leads to a difference of about 4 pK units in 50% Me2SO
and about 5 pKa units in 90% Me2SO between the weakest (2-
Cl) and strongest (2-NO2) acid. Figure 1 shows Hammett plots
of log KCa H versus σ-.19 They yield F values of 4.59 ( 0.41 in
90% Me2SO and 3.77 ( 0.24 in 50% Me2SO. The fact that
correlation with σ- is quite satisfactory and plots versus the
standard σ-values show strong positive deviations for CN, SO2-
CH3 and especially NO2 indicates that resonance effects in the
stabilization of 2-X- are quite strong.
The ratio 50r90 ) F90/F50 of 1.22 ( 0.18 is a measure of
relative sensitivity of pKCa H to X in the two solvents. An
alternate source for 50r90 is the slope of a plot of log KCa H(90)
versus log KCa H(50) (Figure 4); it yields 50r90 ) 1.23 ( 0.04.
Because it is not dependent on the choice of appropriate σ-
values, this second method is superior to the first, as reflected
positive log 0γC90- that C- is less strongly solvated in 90% Me2-
SO than in water.
Discussion
Solvent Dependence of pKCa H. The pKaCH values for all 2-X
decrease with increasing Me2SO content of the solvent. This
is because the stronger solvation of 2-X by Me2SO than by water
is matched by an even somewhat larger increase in the solvation
of the anion 2-X- in the Me2SO-containing solvents, i.e., log0
γ5C0-, log50γ90-, etc. (Table 3) are somewhat more negative than
C
log0γC50H, log50γ90 , etc. (Table 2). Coupled with the signifi-
CH
cantly stronger solvation of the hydronium ion by Me2SO (log0
γ5H0 ) -1.93, log50γH90 ) -1.12)16 this leads to a significant
+
+
CH
decrease in pKCa H as seen, e.g., by solving eq 3 for ∆50pKa
.
0
The stronger solvation of 2-X- by Me2SO is a reflection of
the dispersion of the negative charge, as is also observed for
the conjugate anions of 9-cyanofluorene, 9-carbomethoxyfluo-
(16) Wells, C. F. In Thermodynamic BehaVior of Electrolytes in Mixed
SolutionssII; Furter, W. F., Ed.; American Chemical Society: Washington,
D. C.; Advances in Chemistry 177, p 53.
(18) For example, pKCa H of CH3NO2 is 10.28 in water, 11.32 in 50%
DMSO and 14.80 in 90% DMSO.5
(19) σ- ) 1.25 (NO2), 1.05 (SO2CH3), 0.99 (CN), 0.64 (CF3), 0.26 (Br),
0.24 (Cl).20a
(17) Parker, A. J. Chem. ReV. 1969, 69, 1.

Deprotonation of 2-Nitro-4-X-phenylacetonitriles
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11449
Table 4. Brønsted âB (RBH) and log ko for the Reactions of 2-X
with Piperidine and Morpholine at 20 °Ca
pKCa H
âB
90% Me2SO-10% Water
RBH
log ko
pxy ) ∂âB/∂pKCH
b
+
2-X
a
2-NO2
2-SO2CH3
2-CN
2-CF3
2-Br
6.00
7.85
8.15
9.24
11.03
11.15
0.54
0.53
0.52
0.56
0.64
0.65
0.46
0.47
0.48
0.44
0.36
0.35
2.54
3.18
3.24
3.43
3.79
3.81
0.046 ( 0.012
0.090 ( 0.012
}
2-Cl
50% Me2SO-10% Water
2-NO2
2-SO2CH3
2-CN
2-CF3
2-Br
8.06
9.40
9.53
10.71
12.02
12.17
0.49
0.46
0.46
0.51
0.62
0.63
0.51
0.54
0.54
0.50
0.38
0.37
2.76
3.14
3.15
3.35
3.60
3.62
}
2-Cl
Water
2-NO2
2-SO2CH3
2-CN
9.91
11.12
11.43
0.39
0.40
0.44
0.61
0.60
0.56
2.85
3.14
3.20
Figure 1. Hammett plots of log KCa H vs. σ-. Open circles: 90% Me2-
SO, filled circles: 50% Me2SO.
b
a Estimated error in âB and RBH is (0.02. RBH ) dlog
+
+
BH
k
/dlog KBa H
.
in the much smaller standard deviation. A similar plot (not
shown) of log KaCH(50) versus log KCa H(0) for X ) NO2, SO2-
CH3 and CN yields 50r90 ) 1.01 ( 0.11. The sensitivity of p
KCa H to the para substituent is therefore 1.00, 1.01 and 1.24
-1
(0r50
×
50r90 ) in water, 50% Me2SO and 90% Me2SO,
respectively. The higher sensitivity to the substituent effect in
the less aqueous, i.e., less polar solvent is a common phenom-
enon for ionization reactions,20b but the r values indicate that
the solvent dependence is quite modest. This contrasts with
0r100 ) F(Me2SO)/F(H2O) ≈ 37 for the pKCa H of XC6H4CH2-
NO2 which implies a very strong solvent dependence. This
contrast can be traced to the fact that the 2-X- anions are, as
indicated earlier, better solvated by Me2SO than by water
(negative log0γC50-, and log50γ90-, Table 3) while the opposite is
C
true for PhCHdNO2 (positive log0γC-, and log50γ90-, Table
50
-
C
3).
Brønsted Parameters. From two-point Brønsted plots of
log(k1B/q)21 vs pKaBH + log(p/q)21 (not shown) approximate
Brønsted âB and log ko (log kB/q when pKBH - pKaCH + log-
1
a
(p/q) ) 0) values can be obtained. They are summarized in
Table 4.
Figure 2. Plot of log KCa H(90) versus log KaCH(50).
By plotting log k1B vs log KCa H one may also evaluate
Brønsted RCH values. Such plots are displayed in Figures 3
and 4 for the reactions in 90% and 50% Me2SO, respectively.
These plots show downward curvature which, in principle, could
be interpreted as “Marcus curvature.”22,23 However, we prefer
an interpretation according to which the points for the π-acceptor
substituents CN, SO2CH3 and NO2 are considered to deviate
negatively from the Brønsted line defined by the other substit-
uents (CF3,24 Br, Cl). This sort of negative deviation has been
observed in other systems25 and has been attributed to a lag in
the development of the resonance effect of these substituents
behind proton transfer at the transition state.2a,b This point will
be further discussed under “Substituent Dependence of Intrinsic
Rate Constants.” The RCH values reported in Table 5 are
therefore based on the three non-π-acceptor substituents only,
and no RCH is reported in water since rate constants for the
deprotonation of 2-CF3, 2-Br, and 2-Cl could not be measured
in this solvent.
Changes in Transition State Structure. RCH decreases with
increasing pKBa H while âB for 2-CF3, 2-Cl and 2-Br increases
with increasing pKCa H. These changes in the Brønsted coef-
(20) (a) Exner, O. In Correlation Analysis in Chemistry, Chapman, N.
B.; Shorter, J., Eds.; Plenum: New York, 1978; p 439. (b) Exner, O. In
AdVances in Linear Free Energy Relationships, Chapman, N. B.; Shorter,
J., Eds.; Plenum: New York, 1972; p 1.
(21) p ) 2 and q ) 1 are statistical factors referring to the number of
equivalent protons on BH+ and the number of basic sites on B, respectively.
(22) (a) Marcus, R. A. J. Phys. Chem. 1968, 72, 891. (b) Cohen, A. D.;
Marcus, R. A. Ibid. 1968, 72, 4249.
ficients are related by the interaction coefficient pxy ) ∂âB/∂p
KCa H ) ∂RCH/-∂pKBH
.
Plots of âB vs pKCH (Figure 3S)14
23,26
a
yield ∂âB/∂pKCH ) a0.046 ( 0.012 in 90% Me2SO and 0.090 (
a
0.012 in 50% Me2SO while ∂RCH/-∂pKBH is 0.049 ( 0.011 in
90% Me2SO and 0.086 ( 0.017 in 50% Me2SO.
a
(23) Jencks, W. P. Chem. ReV. 1985, 85, 511.
(24) Strictly speaking CF3 is also a π-acceptor but it is quite weak (σ-
) 0.64 vs σ ) 0.54)20a and is, for the purposes of our discussion, included
with the substituents that are not π-acceptors.
The positive pxy values reflect changes in transition state
structure with changing pKa of carbon acid and amine, respec-
(25) (a) Bunting, J. W.; Stefanidis, D. J. Am. Chem. Soc. 1988, 110,
4008. (b) Stefanidis, D.; Bunting, J. W. J. Am. Chem. Soc. 1991, 113, 991.
(26) Jencks, D. A.; Jencks, W. P. J. Am. Chem. Soc. 1977, 99, 7948.

11450 J. Am. Chem. Soc., Vol. 118, No. 46, 1996
Bernasconi and Wenzel
-
Table 5. Brønsted RCH (âC ) for the Reactions of 2-X (X ) CF3,
Br, Cl) with Piperidine and Morpholine at 20 °C
pKBa H
RCH
90% Me2SO-10% Water
âC
pxy ) ∂RCH/-∂pKBH
a
-
R2NH
a
piperidine 10.74 0.36 ( 0.01 0.64 ( 0.01
morpholine 8.91 0.45 ( 0.01 0.52 ( 0.01
0.049 ( 0.011
0.086 ( 0.017
50% Me2SO-50% Water
piperidine 11.05 0.38 ( 0.02 0.62 ( 0.02
morpholine 8.72 0.58 ( 0.02 0.42 ( 0.02
BH
-1
a âC ) dlog k /dpKCa H
.
-
Imbalance. The deprotonation of carbon acids that lead to
resonance stabilized carbanions is characterized by transition
state imbalances which manifest themselves in RCH not being
equal to âB.2a,b These imbalances are caused by a lag in the
development of resonance and of the solvation of the developing
carbanion behind proton transfer. This can lead to either RCH
> âB or RCH < âB. The former situation prevails when the
transformation of the transition state into the carbanion leads
to a shift of the negative charge away from the X-substituent
(e.g., XC6H4CH2NO2 + R2NH, eq 5, RCH ) 1.29,4 âB ) 0.485
in water), while in the latter situation negative charge shifts
toward the X-substituent (e.g., 4-X + R2NH, eq 6, RCH ) 0.28,
âB ) 0.47 in 50% Me2SO).28 This interpretation of RCH and
âB presumes that âB can be viewed as at least an approximate
Figure 3. Brønsted plots of log kB1 versus log KaCH in 90% Me2SO.
Open symbols: piperidine; filled symbols: morpholine; circles: 2-Cl,
2-Br and 2-CF3; squares: 2-CN, 2-SO2CH3 and 2-NO2.
measure of proton or charge transfer at the transition state,29
while RCH is clearly not such a measure because it is distorted
by the lag in the resonance development or charge delocaliza-
tion/solvation. Note that the magnitude of |RCH - âB| tends to
be large when the carbanion is strongly stabilized by resonance
and solvation (|RCH - âB| ) 0.81 for ArCH2NO2 + R2NH in
water) and smaller for less strongly stabilized carbanions (|RCH
- âB| ) 0.19 for 4-X + R2NH in 50% Me2SO).
Figure 4. Brønsted plots of log kB1 versus log KaCH in 50% Me2SO.
Open symbols: piperidine; filled symbols: morpholine; circles: 2-Cl,
2-Br and 2-CF3; squares: 2-CN, 2-SO2CH3 and 2-NO2.
tively. Other systems for which positive pxy values have been
reported are the deprotonation of 3-X by primary aliphatic
amines (0.04) and carboxylate ions (0.07) in water,27 and the
deprotonation of 4-X by primary aliphatic amines (0.03) and
by the piperidine/morpholine pair (0.01) in 50% Me2SO-50%
The reaction of 2-X with amines also shows an imbalance.
Because of the dependence of RCH on pKBa H, and of âB on p
KCa H, we shall use average Brønsted coefficients (RjCH, âhB), as
for the reactions of 4-X.28 Furthermore, since âB is likely to
be influenced by the resonance effect of the π-acceptors p-NO2,
p-SO2CH3 and p-CN, âhB is based on 2-CF3, 2-Br and 2-Cl only.
In 90% Me2SO, RjCH ) 0.40 and âhB ) 0.62, for an imbalance
RjCH - âhB ) -0.22; in 50% Me2SO, RjCH ) 0.48 and âhB
)
0.58, for an imbalance RjCH - âhB ) -0.10.
In view of the fact that averages of RCH and âBH are used,
and that each Brønsted coefficient is associated with an
uncertainty in the order of (0.02, the numerical values of these
imbalances should be regarded as crude estimates from which
water.28 These changes can be understood on the basis of a
reaction coordinate diagram as described by Murray and
Jencks27 and also discussed by Bernasconi and Fairchild;28 these
papers should be consulted for details.
(29) This is the traditional view23,30 although this view has been
challenged.31 The fact that one finds RCH > âB in cases where the charge
in the carbanion presumably shifts away from the X-substituent whereas
one observes RCH < âB in cases where the charge in the carbanion is
presumably closer to the X-substituent than it is in the transition state may
itself be taken as strong evidence in support of the traditional view.
(27) Murray, C. J.; Jencks, W. P. J. Am. Chem. Soc. 1990, 112, 1880.
(28) Bernasconi, C. F.; Fairchild, D. E. J. Phys. Org. Chem. 1992, 5,
409.

Deprotonation of 2-Nitro-4-X-phenylacetonitriles
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11451
Table 6. Solvent Effects on log ko of the Reactions of Carbon
Acids with the Piperidine/Morpholine Pair at 20 °C
only qualitative conclusions can be drawn. Nevertheless, an
alternate method for estimating RjCH - âhB discussed below yields
values that are consistent with the ones reported here.
CH
0δ50log ko
50δ90log ko
0δ90log ko
The qualitative conclusions are (a) that RjCH - âhB is a negative
number and (b) the absolute value of the imbalance is larger in
90% Me2SO than in 50% Me2SO. The negative sign is
consistent with charge shifting closer to the X-substituent during
the transformation of the transition state into the carbanion (eq
7), a situation which is similar to that for eq 6. The larger
2-NO2
2-SO2CH3
2-CN
2-CF3
2-Br
-0.09
0.00
-0.05
-0.22
0.04
0.09
0.08
0.19
-0.31
0.04
0.04
2-Cl
0.19
9-CN-Fla,b
[(CO)3C6H5]2CHCNc
9-COOMe-Fla,b
1,3-indandioned
0.14g,h
-0.19g
-0.32
0.25g
0.72
-0.05g,i
0.16h
0.15
0.97
1.32
0.88i
1.04
2.97
3.65
e
CH2(COCH3)2
PhCH2NO2
CH3NO2
0.89
2.00
2.33
f
f
a Fl ) fluorene. b Reference 11. c (R-Cyanodiphenylmethane)bis(tri-
carboxylchromium(0)), ref 12. d Reference 10. e Reference 8. f Refer-
ence 5. g Reactions with primary aliphatic amines. h 10δ50log ko. i 10δ90log
ko.
magnitude of the imbalance in the Me2SO-rich solvent is mainly
a consequence of the stronger solvation of 2-X- by Me2SO than
by water, as seen in the negative log50γC90 values (Table 3).
-
This contrasts with a smaller imbalance in Me2SO than in water
for the reaction of XC4H6CH2NO2 with amines (RCH - âB )
0.90 - 0.55 ) 0.35 in Me2SO;6 RCH - âB ) 1.29 - 0.48 )
0.81 in water32), which is a consequence of the weaker solvation
resonance effect which is a consequence of the lag in the
delocalization of the negative charge into X behind proton
transfer at the transition state.34 This is the same effect
that leads to the negative deviations from the Brønsted plots
(Figures 3 and 4) discussed earlier. Hence, it is not surprising
that with the strongest π-acceptor, NO2, log ko is particularly
low.
of XC6H4CH)NO2 (positive log γ9C0-, Table 3) by Me2SO
than by water.
-
0
Substituent Dependence of Intrinsic Rate Constants. The
intrinsic rate constants are summarized as log ko in Table 4.
They show a decreasing trend with increasing electron with-
drawing strength of the X-substituent. For the substituents that
only exert an inductive effect (X ) Cl, Br, CF3) the dependence
of log ko on X is relatively small and is simply a consequence
of the imbalanced transition state. It can be shown that the
Solvent Effects on Intrinsic Rate Constants. Table 6
provides a summary of the solvent effects on log ko for the
deprotonation of 2-X and several other carbon acids by the
piperidine/morpholine pair. 0δ50 log ko is defined as log ko(50%
Me2SO) - log ko(water), 50 90 log ko as log ko(90% Me2SO) -
δ
log ko(50% Me2SO), etc. These solvent effects fall into three
distinct categories. The first comprises all the 2-X as well as
(R-cyanodiphenylmethane)bis(tricarbonylchromium(0)), 9-cy-
ano- and 9-carbomethoxyfluorene. Their log ko values show a
small solvent dependence and, in some cases, tend to be slightly
lower in the less aqueous solvent. The second group consists
of the two â-diketones; their log ko values increase by roughly
one unit from water to 90% Me2SO, or somewhat less than one
unit from 50% to 90% Me2SO. The third group, the nitroal-
kanes, show qualitatively similar behavior as the diketones but
the solvent effects are much larger.
change in log ko with changing pKCH is given by eq 82a,b (I )
a
inductive effect) for the ideal case where pxy ) 0; for pxy * 0
eq 8
δlog kIo
-δpKCa H
) (RCH - âB)
(8)
should be a good approximation with RCH replaced by RjCH, âB
by âhB.33
Plots of log ko vs pKCH (not shown) afford RjCH - âhB
)
a
To a good approximation, these solvent effects can be
explained by invoking two main factors. The first is the
difference in the solvation of the carbanion in the different
solvents, and the fact that, at the transition state, the solvation
of the incipient carbanion lags behind proton transfer. When
solvation of the carbanion is weaker in the Me2SO-richer
solvents (log0γC50 or log0γ9C0 > 0), as in the case of the
-0.20 ( 0.01 in 90% Me2SO and -0.18 ( 0.02 in 50%
Me2SO; the value in 90% Me2SO is close to RjCH - âhB
)
-0.22 obtained directly from RjCH and âhB (see previous section)
but the value in 50% Me2SO is not in close agreement with
that calculated directly (-0.10). The relatively poor agree-
ment between RjCH - âhB obtained via eq 8 and that calculated
directly in 50% Me2SO probably reflects the relatively large
experimental uncertainties in RjCH and âhB mentioned pre-
viously, the large difference between RjCH and the actual
RCH values, and the likelihood that the standard deviations
associated with RjCH - âhB obtained via eq 8 underestimate the
true error.
-
-
nitroalkanes and â-diketones, the solvational lag enhances log
ko; when the solvation of the carbanion is stronger in the Me2-
SO-richer solvent (log0γC50 or log0γ9C0 < 0), as is the case for
-
-
all 2-X, the fluorene derivatives and (R-cyanodiphenylmethane)-
bis(tricarbonylchromium(0)), the solvational lag decreases log
ko. The second factor is a “classical” solvent effect2b,35 which
is best understood in terms of preferential solvation of the
polarizable transition state by the polarizable Me2SO. It
increases log ko and hence reinforces the log ko enhancing effect
of the first factor in the reactions of the nitroalkanes and
â-diketones but more or less offsets the log ko reducing effect
of the first factor in the reactions of the 2-X, the fluorene
For the π-acceptor substituents (CN, SO2CH3, NO2) the ko-
reducing inductive effect is augmented by a ko-lowering
(30) (a) Leffler, J. E.; Grunwald, E. Rates and Equilibria of Organic
Reactions, Wiley: New York, 1963; p 156. (b) Kresge, A. J. Acc. Chem.
Res. 1975, 8, 354.
(31) Pross, A. J. Org. Chem. 1984, 49, 1811. (b) Bordwell, F. G.; Hughes,
D. L. J. Am. Chem. Soc. 1985, 107, 4737. (c) Pross, A.; Shaik, S. S. New
J. Chem. 1989, 13, 427.
(32) Based on data from reference 4.
(34) Note that this effect is on top of that of the delayed delocalization
into the ortho-nitro group.
(35) See also Gandler, J. R.; Bernasconi, C. F. J. Am. Chem. Soc. 1992,
114, 631.
(33) Note that in a reaction such as the deprotonation of arylnitromethanes
where RCH - âB > 0, eq 8 indicates that log ko increases with increasing
electron withdrawing strength of the phenyl substituent, as observed.2a,2b

11452 J. Am. Chem. Soc., Vol. 118, No. 46, 1996
Bernasconi and Wenzel
Table 7. Summary of Melting Points, Yields, and Proton NMR
Data for 2-X
derivatives and (R-cyanodiphenylmethane)bis(tricarbonyl-
chromium(0)).
Our results show that the hypothesis formulated in the
Introduction according to which the solvent effect on ko for the
reactions of 2-NO2 and the other 2-X should be intermediate
between the solvent effects on ko for deprotonation of acety-
lacetone and 9-cyanofluroene is not confirmed. This hypothesis
NMR chemical shiftsa
2-X
mp (°C) yield 3-position 5-position 6-position CH2
b
2-SO2CH3 120-121 ≈90% 8.73 (1.8) 8.28 (8.2; 1.8) 8.03 (8.2) 4.32
2-CN
2-CF3
2-Br
94-95 ≈40% 8.48 (1.6) 8.02 (8.1; 1.6) 7.95 (8.1) 4.31
38-39 ≈70% 8.66 (1.6) 8.12 (8.1; 1.6) 7.98 (8.1) 4.29
109-110 ≈40% 8.31 (2.0) 7.84 (8.3; 2.0) 7.62 (8.3) 4.16
84-85 ≈35% 8.31 (1.9) 7.78 (8.3; 2.0) 7.71 (8.3) 4.19
-
was based on the expectation that the solvation of 2-NO2 in
2-Cl
water should include some hydrogen bonding to the partially
negative nitro groups (or the ortho-nitro group in 2-X with X
* NO2) which would lead to log γC-values between those for
9-cyanofluorenyl anion (log0γC90 ) -4.02, log50γ9C0 ) -2.65)
a All spectra taken in CDCl3 at 250 MHz. Numbers in parentheses
are coupling constants in Hz; CH2 protons give singlet. b Methyl group
gives singlet at 3.15 ppm.
-
-
and acetylacetonate anion (log0γC90 ) 5.03, log50γC90 ) 2.36).
-
-
(2-Nitro-4-trifluoromethyl)phenylacetonitrile (2-CF3). 4-Chloro-
3-nitrobenzotrifluoride (Aldrich) was first distilled under reduced
pressure. Reflux time of the mixture of NCChHCOOEt with ArCl was
8 h, and 4 h for the ethanolic HCl solution of ArCH(CN)COOEt. Upon
neutralization and cooling of the ethanolic HCl solution the product
appeared as an oil which was steam distilled and then recrystallized
from 80% ethanol/20% water several times. HPLC showed a single
peak.
(4-Cyano-2-nitro)phenylacetonitrile (2-CN). 4-Chloro-3-nitroben-
zonitrile (ICN Chemicals) was used without purification. Reflux times
were 8 h and 8 h. The product was extracted from the ethanolic solution
into methylene chloride. A yellow oil was obtained after evaporation
of the extract which subsequently solidified. The solid was recrystal-
lized twice from low boiling (30-60 °C) petroleum ether. HPLC
showed a single peak.
(4-Bromo-2-nitro)phenylacetonitrile (2-Br). 2,5-Dibromonitroben-
zene (Eastman Kodak) was recrystallized from 80% ethanol prior to
use. To 0.25 mol of NCChHCOOEt in 250 mL 0.02 mol of 2,5-
dibromonitrobenzene was added. After 6 h reflux an additional 0.25
mol of NCChHCOOEt was added and refluxed for another 6 h; without
the second reflux much of the 2,5-dibromonitrobenzene remained
unreacted. Reflux time in ethanolic HCl was 6 h. The product was
extracted into methylene chloride which yielded a brown oil after rotary
evaporation. The oil was taken up in boiling ethanol and decolorized
with charcoal. Water was then added to bring about a 80% ethanol/
20% water mixture; crystallization took several days. The product was
recrystallized from 80% ethanol/20% water.
(4-Chloro-2-nitro)phenylacetonitrile (2-Cl). 2,5-Dichloronitroben-
zene (Eastman Kodak) was recrystallized from 80% ethanol prior to
use. All procedures were the same as for the synthesis of 2-Br.
Buffers and Other Reagents. Reagent grade piperidine and
morpholine from Aldrich were refluxed over CaH2, distilled and stored
over P2O5 prior to use. Benzoic acid (Aldrich) was recrystallized from
ethanol and stored over P2O5 prior to use. Chloroacetic acid (Mallinck-
rodt) was recrystallized from CCl4 prior to use. The last three
compounds were used as buffers to calibrate the pH-meter in DMSO/
water mixture according to Halle´ et al.40
90
90
-
However, log0γC- ) -2.54, and log50γC- ) -1.93 for 2-NO2
are much closer to the corresponding values for 9-cyanofluorenyl
anion than to those of the acetylacetonate anion. Apparently,
Me2SO is a better solvator than water for nitro groups that carry
only a small fraction of charge (in contrast to a nitro group
with a full negative charge as in RCHdNO2-). This conclusion
is consistent with findings by Keeffe et al.7 and others.36,37
A more quantitative treatment, based on a recently proposed
formalism,38 that takes other factors influencing the solvent
effects on log ko into account is presented under Supporting
Information.14 These factors include delayed solvation of the
developing ammonium ion (protonated amine) and desolvation
of the amine and carbon acid that is ahead of proton transfer.
This quantitative treatment confirms the above qualitative
conclusions.
Experimental Section
Materials. The synthesis of the various 4-X-2-nitrophenylacetoni-
triles was based on the method of Fairbourne and Fawson,39 eq 9.
NaOEt was generated by dissolving Na in ethanol.
NCCH2COOEt NaOEt8 NCChHCOOEt ArCl(Br)8
EtOH
ArCH(CN)COOEt HCl8 ArCH2CN (9)
The sodium salt of the ethylcyanoacetate anion was prepared as an
approximately 0.5 M solution in ethanol by adding one equivalent of
NaOEt to an ethanolic solution of ethylcyanoacetate (Aldrich). After
adding 0.1 equivalents of neat ArCl to 100 mL of the above solution,
the reaction mixtures were refluxed for up to 12 hours, acidified with
nitric acid and extracted with methylene chloride. Evaporation of the
solvent yielded ArCH(CN)COOEt as a reddish-brown oil in most cases
although with Ar ) 2-nitrophenyl-4-trifluoromethyl and 4-methylsul-
fonyl-2-nitrophenyl spontaneous crystallization occurred. The (4-X-
2-nitrophenyl)ethylcyanoacetates were then added to a boiling ethanolic
3 M HCl solution and transformed to the 4-X-2-nitrophenylacetonitriles
by refluxing for 4-6 h. Table 7 summarizes yields, melting points
Reagent grade KCl (Mallinckrodt) was used without further purifica-
tion. Stock solutions of HCl and KOH, used to adjust the pH of
solutions, were prepared from “Dilut It” (Baker Analytical) stock
solutions. DMSO (Mallinckrodt) was distilled from CaH2 under
reduced pressure; the DMSO was generally used within one week of
distillation.
1
and H NMR data; the latter were taken in CDCl3 on a Brucker 250
MHz instrument. Further details were as follows.
2-4-Dinitrophenylacetonitrile (2-NO2). This compound was avail-
able from a previous study13 and recrystallized from CCl4 prior to use,
mp 88-89 °C (lit. mp 89 °C39).
(4-Methylsulfonyl-2-nitro)phenylacetonitrile (2-SO2CH3). Reflux
time of the mixture of NCChHCOOEt with 4-chloro-3-nitromethylsul-
fonylbenzene and of ArCH(CN)COOEt with ethanolic HCl was for 4
h. After neutralization of the ethanolic HCl solution, 2-SO2CH3 formed
crystals which were recrystallized several times from a 50% ethanol/
50% water mixture.
Solutions and pH-Measurements. Solutions were prepared and
their pH measured as described41 before.
UV Spectra and Spectrophotometric pKCa H Determinations.
The anions 2-X- absorb strongly in the visible region of the spectrum
whereas 2-X do not. Spectra were taken in a Hewlett-Packard diode
array spectrophotometer; λmas and ꢀ values are summarized in Table 8.
The pKCa H values were determined by applying eq 10 where A, AC-
and ACH (≈ 0) are the absorbances at pH ∼ pKCH, pH . pKaCH and pH
a
(40) Halle´, J.-C.; Gaboriaud, R.; Schaal, R. Bull. Soc. Chim. Fr. 1970,
2047.
(36) (a) Fujio, M.; McIver, R. I.; Taft, R. W. J. Am. Chem. Soc. 1981,
103, 4017. (b) Taft, R. W. Prog. Phys. Org. Chem. 1983, 14, 249.
(37) Mashima, M.; McIver, R. I.; Taft, R. W.; Bordwell, F. G.; Olmstead,
W. N. J. Am. Chem. Soc. 1984, 106, 2717.
(38) The most recent version of this formalism which differs somewhat
from previous treatments2a,11,12 is described in reference 2b.
(39) Fairbourne, A.; Fawson, H. R. J. Chem. Soc. 1927, 46.
(41) (a) Bernasconi, C. F.; Hibdon, S. A. J. Am. Chem. Soc. 1983, 105,
4343. (b) Bernasconi, C. F.; Bunnell, R. D. Isr. J. Chem. 1985, 26, 420. (c)
Bernasconi, C. F.; Paschalis, P. J. Am. Chem. Soc. 1986, 108, 2969. (d)
Bernasconi, C. F.; Terrier, F. J. Am. Chem. Soc. 1987, 109, 7115. (e)
Bernasconi, C. F.; Stronach, M. W. J. Am. Chem. Soc. 1990, 112, 8448.
(42) The average of the âB values in the two solvents is used.

Deprotonation of 2-Nitro-4-X-phenylacetonitriles
J. Am. Chem. Soc., Vol. 118, No. 46, 1996 11453
Table 8. Summary of UV/vis Spectral Data
direction 2-X + B f 2-X- + BH+, 2-X was dissolved in 0.001 M
HCl while the buffers were prepared with an excess of 0.001 M KOH.
For reactions conducted in the direction 2-X- + BH+ f 2-X + B,
2-X was dissolved in 0.001 M KOH while the buffers were prepared
with an excess of 0.001 M HCl. All rates were measured by monitoring
λmax (ꢀ)
2-X
90% DMSO
50% DMSO
water
2-NO2
2-SO2CH3
2-CN
2-CF3
2-Br
442 (26 700)
560 (9 400)
570 (9 800)
570 (8 500)
589 (10 900)
589 (10 400)
442 (15 100)
560 (10 000)
580 (9 000)
570 (11 000)
589 (9 900)
589 (9 600)
442 (19 800)
475 (9 800)
580 (9 000)
the appearance or disappearance of 2-X- at or near its λmax
.
Partition Coefficients. The procedures described earlier9 were
followed except that p-xylene instead of n-heptane was used as the
partition solvent; none of the 2-X were soluble enough in n-heptane to
give meaningful results.
2-Cl
Acknowledgment. This research was supported by Grant
No. CHE-9307659 from the National Science Foundation.
, pKCa H, respectively.
pKCa H ) pH + log
Supporting Information Available: Figures S1 and S2,
kinetic data; Tables S1 and S2, estimates of various contributions
to solvent effects; and text, semi-quantitative treatment of solvent
effects on intrinsic rate constants (10 pages). See any current
masthead page for ordering and Internet access instructions.
AC - A
A - ACH
(10)
Kinetic Measurements. All kinetic measurements were performed
on a Durrum-Gibson stopped-flow spectrophotometer. For runs in the
JA961837Q