Kinetics and Mechanism of the Addition of Benzylamines to Benzylidenemalononitriles in Acetonitrile

Article abstract of DOI:10.1021/jo991823d

Nucleophilic addition reactions of benzylamines (BA) to benzylidenemalononitrile (BMN) have been studied in acetonitrile at 15.0°C. The rate is first-order with respect to both BA and BMN and no base catalysis is observed. The rate decreases as the electr

Full text of DOI:10.1021/jo991823d

2188
J . Org. Chem. 2000, 65, 2188-2191
Kin etics a n d Mech a n ism of th e Ad d ition of Ben zyla m in es to
Ben zylid en em a lon on itr iles in Aceton itr ile
Hyuck Keun Oh, J in Hee Yang, Hai Whang Lee,† and Ikchoon Lee*,†
Department of Chemistry, Chonbuk National University, Chonju, 560-756 Korea, and
Department of Chemistry, Inha University, Inchon, 402-751 Korea
Received November 29, 1999
Nucleophilic addition reactions of benzylamines (BA) to benzylidenemalononitrile (BMN) have been
studied in acetonitrile at 15.0 °C. The rate is first-order with respect to both BA and BMN and no
base catalysis is observed. The rate decreases as the electron-withdrawing power of the substituent
(Y) in the substrate increases (FY < 0). This is in contrast to the similar reactions of â-nitrostyrenes
(NS) with BAs in acetonitrile and the addition reactions of NS and BMN in aqueous solution (FY >
0). This sign change of FY is considered to result from the strong electron-withdrawing power of
the (CN)2 group in BMN, which leads to polarization of CRδ+Cδ-(CN)2 in the transition state. The
mechanism of amine addition to BMN in acetonitrile is radically different from that in water. The
reaction is predicted to proceed concertedly in a single step with a hydrogen-bonded, four-center
cyclic transition state.
In tr od u ction
weakly resonance stabilized. Their work was carried out
in aqueous solution, since delocalization of the negative
charge into the activating groups (Z ) Z′ ) CN) is
strongly dependent on the solvation of this negative
charge.2,3 They found that even though the reactions
show the typical behavior of a carbanion-forming process,
the carbanion is largely stabilized by polar effects while
resonance effects playing a more modest role.
On the other hand, our recent work on the nucleophilic
addition reactions of benzylamines (BA) to â-nitrosty-
renes (NS) in acetonitrile4 showed that proton transfer
to the â-carbon occurs concurrently and the neutral
adducts are formed in a single step, not through a
zwitterionic addition intermediate, T( (of the type in eq
1). We attributed this change of mechanism in aceto-
nitrile to (i) weak solvation by acetonitrile to stabilize
the carbanion (and benzylammonium ion) in the putative
intermediate and (ii) hydrogen bonding to negative
charge localized on Câ in the TS due partly to the well-
known “imbalance”, which causes a lag in charge delo-
calization into the NO2 moiety behind C-N bond forma-
tion.
Addition of anionic or neutral base nucleophiles to
olefins is an important class of reaction. It may occur in
a nucleophilic vinylic substitution1 or in an adduct
formation.2 In any of these classes of reactions, the role
of an activating group is similar. The activating groups
are normally electron acceptors, which act to stabilize the
intermediate carbanion in the stepwise pathways.1 The
addition reactions of amine nucleophiles to olefins con-
taining activating groups have been extensively studied.2
One of the goals of such studies has been to demonstrate
the transition state “imbalance” in which a lag in the
charge delocalization into the activating group behind
C-Nu bond formation occurs in the transition state. This
type of “imbalance” arises from a delayed development
of product (or carbanionic intermediate) stabilizing reso-
nance in the transition state, in which a delayed solvation
of the carbanionic center plays an important role espe-
cially in aqueous solution.3
Nucleophilic addition reactions of amine bases to
substituted benzylidenemalononitriles (BMN) have been
studied by Bernasconi et al.,2 with the view of demon-
strating a transition state (TS) imbalance2,3 involved in
a carbanion-forming reaction (eq 1, where Nu ) second-
In view of this surprising mechanistic change observed
in the additions of amines to â-nitrostyrenes (NS) in
acetonitrile, we extend this series of work to confirm such
a change of mechanism using carbanion-forming reac-
tions involving relatively weak activating groups, Z ) Z′
) CN, in acetonitrile. The resonance stabilization of the
carbanion is reported to be much weaker with the
activating groups of (CN)2 than with H(NO2).2,3
Another point of interest is the sign and magnitude of
ary amines and Z ) Z′ ) CN) in which the carbanion is
the cross-interaction constant, FXY in eqs 24
* To whom corespondence should be addressed. E-mail: ilee@
dragon.inha.ac.kr. Fax: +82-32-865-4855.
log(kXY/kHH) ) FXσX + FYσY + FXYσXσY
FXY ) ∂FX/∂σY ) ∂FY/σX
(2a)
(2b)
(1) (a) Apeloig, Y.; Rappoport, Z. J . Am. Chem. Soc. 1979, 101, 5095.
(b) Rappoport, Z. Acc. Chem. Res. 1981, 14, 7. (c) Cohen, D.; Bar, R.;
Shaik, S. S. J . Am. Chem. Soc. 1986, 108, 231. (d) Bernasconi, C. F.;
Schuck, D. F.; Ketner, R. J .; Weiss, M.; Rappoport, Z. J . Am. Chem.
Soc. 1994, 116, 11764. (e) Beit-Yannai, M.; Rappoport, Z.; Shainyan,
B. A.; Danilevich, Y. S. J . Org. Chem. 1997, 62, 8049. (f) Chen, X.;
Rappoport, Z. J . Org. Chem. 1998, 63, 5684.
where X and Y are substituents in the nucleophile,
benzylamine (BA), and in the substrate, benzylidene-
(2) (a) Bernasconi, C. F.; Killion, R. B., J r. J . Org. Chem. 1989, 54,
2878. (b) Bernasconi, C. F. Tetrahedron 1989, 45, 4017.
(3) Bernasconi, C. F. Acc. Chem. Res. 1987, 20, 301.
(4) Oh, H. K.; Yang, J . H.; Sung, D. D.; Lee, I. J . Chem. Soc., Perkin
Trans. 2, in press.
10.1021/jo991823d CCC: $19.00 © 2000 American Chemical Society
Published on Web 03/14/2000

Addition of Benzylamines to Benzylidenemalononitriles
J . Org. Chem., Vol. 65, No. 7, 2000 2189
Ta ble 1. Secon d -Or d er Ra te Con sta n ts, k2 × 10 L m ol-1 s-1 for th e Ad d ition of X-Ben zyla m in es to
Y-Ben zylid en em a lon on itr iles in Aceton itr ile a t 15.0 °C
Y
a
X
p-Me
44.4b
H
p-Cl
p-Br
18.7
p-NO2
16.5
11.1
7.38
7.73
3.94
FY
p-OMe
-0.47 ( 0.02
30.7
25.7
19.3
14.6
20.9c
p-Me
H
p-Cl
23.9
13.3
8.31
5.68
19.2
10.1
13.9
7.21
-0.52 ( 0.02
-0.55 ( 0.03
-0.63 ( 0.04
7.66
2.04
1.40
4.12
2.92
2.77
3.82
-1.49 ( 0.07
1.42 ( 0.10
0.943
-1.81 ( 0.07
1.73 ( 0.10
d
FX
-1.62 ( 0.06
1.54 ( 0.10
-1.67 ( 0.08
1.59 ( 0.12
-1.69 ( 0.07
1.60 ( 0.11
FXYe ) -0.31
f
âX
a
b
d
The σ values were taken from ref 6. Correlation coefficients were better than 0.995 in all cases At 25.0 °C. c At 5.0 °C. The source
of σ is the same as for footnote a. Correlation coefficients were better than 0.998 in all cases. e Correlation coefficients was 0.998. f The
pKa values were taken from Fischer, A.; Galloway W. J .; Vaughan, J . J . Chem. Soc. 1964, 3588. Oh, H. K.; Lee, J . Y.; Lee, I. Bull. Korean
Chem. Soc. 1998, 19, 1198. Correlation coefficients were better than 0.995 in all cases.
Ta ble 2. Kin etic Isotop e Effects on th e Secon d -Or d er
malononitrile (BMN), respectively, in eq 3. The FXY has
Ra te Con sta n ts (k2) for th e Rea ction s of
Y-Ben zylid en em a lon on itr iles w ith Deu ter a ted
been shown to have a negative sign in addition pro-
cesses,5 e.g., bond formation in the SN2 reactions4a and
X-Ben zyla m in es in Aceton itr ile a t 15.0 °C
in the addition of amines to a vinyl carbon4 with ca. -0.6
X
Y
kH/10-1 M-1 s-1 kD/10-1 M-1 s-1
kH/kD
to -0.8.
p-OMe p-Me
p-OMe
p-OMe p-Cl
p-OMe p-Br
3.07 ( 0.03
2.57 ( 0.02
1.93 ( 0.02
1.87 ( 0.02
1.30 ( 0.01
2.36 ( 0.03a
2.45 ( 0.03
2.55 ( 0.04
2.57 ( 0.03
2.71 ( 0.04
2.25 ( 0.03
2.38 ( 0.04
2.47 ( 0.03
2.56 ( 0.03
H
1.05 ( 0.01
0.757 ( 0.007
0.728 ( 0.005
0.410 ( 0.004
p-OMe p-NO2 1.11 ( 0.01
p-Cl
p-Cl
p-Cl
p-Cl
p-Cl
p-Me
H
p-Cl
p-Br
0.568 ( 0.005 0.253 ( 0.002
0.412 ( 0.004 0.173 ( 0.002
0.292 ( 0.003 0.118 ( 0.001
0.277 ( 0.002 0.108 ( 0.001
p-NO2 0.140 ( 0.001 0.0528 ( 0.0003 2.65 ( 0.03
a
Resu lts a n d Discu ssion
Standard deviations.
The kinetic law found for all the reactions studied in
this work is given by eqs 4 and 5 where k0 () 0) and k2
are the rate constants for solvolysis and aminolysis of
the substrate (BMN), respectively.
in the substrate increases, e.g., k2 (Y ) p-NO2)/k2(Y )
H) = 0.3-0.4. This is opposite to the trend found for the
reactions of NSs4 with BA in acetonitrile [k2(p-NO2)/
k2(H) ) 31-49] and also to that of k1 for the rate-limiting
T( formation of the reactions of BMN with piperidine
(and morpholine) in water [k1(Y ) p-NO2)/k1(Y ) H) =
4].2a Consequently, the sign of FY () -0.47 to -0.63) is
negative for the present reaction series in contrast to the
all positive FY values for other similar reactions compared
above.2 The magnitudes of FX(Fnuc) () -1.49 to -1.81) and
âX(ânuc) () 1.42-1.73) are somewhat greater than those
corresponding values of the NS reactions.4 The negative
FY obtained in this work indicates that the reaction center
carbon (CR) becomes more positive on going from the
reactant to the TS. This is certainly not consistent with
the TS for rate-limiting formation of the addition inter-
mediate, T(, for which negative charge (FY > 0) develops
normally.2-5
The normal kinetic isotope effects (kH/kD > 1.0 in Table
2) involving deuterated benzylamines (XC6H4CH2ND2)
provide evidence for partial N-H(D) bond cleavage in the
TS.7 This means that proton transfer to Câ occurs
concurrently with the N-CR bond formation in the TS,
I, i.e., the reaction does not proceed through an addition
intermediate, T(, but occurs concertedly in a single step.
This mechanism is similar to that proposed for the
addition reactions of benzylamines to NS in acetonitrile.4
We stress that this mechanism is radically different from
the rate-limiting formation of T( in water.2 We propose
again that this mechanistic change can be attributed to
the weak solvation by acetonitrile of the carbanion and
-d[BMN]/dt ) kobs[BMN]
kobs ) ko + k2[BA]
(4)
(5)
No catalysis by a second benzylamine molecule was
detected, which is in contrast to the benzylamine cataly-
sis (kobs ) k2[BA] + k3[BA]2) found for the aminolysis of
â-nitrostyrenes (NSs).4 Plots of kobs vs [BA] were linear,
and the k2 values obtained from the slopes of these plots
are summarized in Table 1. The rate of benzylamine
addition to BMN is faster by more than 102 times than
those to NS, e.g., for X ) Y ) H in acetonitrile, k2 ) 1.01
L mol-1 s-1 at 15.0 °C for BMN, but it is 2.63 × 10-2
L
mol-1 s-1 at 25.0 °C for NS.4 The faster rates observed
with BMN should be due to a stronger electron-deficient
center, CR, in BMN than in NS since both the inductive
(σI) and resonance (σR) electron-withdrawing power of the
two CN groups (2 × σ ) 2 × 0.60 ) 1.20 and 2 × σR )
I
0.36)6 in the BMN is much greater than that of one NO2
group (σI ) 0.65 and σR ) 0.16)6 in the NS. However,
these rates are extremely slower (by a factor of ca.
∼10-5-10-6) than those corresponding reactions in aque-
ous solution.2 We note in Table 1 that the rate decreases
as the electron-withdrawing power of the substituent (Y)
(5) (a) Lee, I. Prog. Phys. Org. Chem. 1992, 27, 57. (b) Lee, I. Chem.
Soc. Rev. 1990, 19, 317. (c) Isaacs, N. S. Physical Organic Chemistry,
2nd ed.; Longman: Harlow, 1995; p 186.
(6) Hansch, C.; Leo, A.; Taft, R. W. Chem. Rev. 1991, 91, 165.
(7) Lee, I. Chem. Soc. Rev. 1995, 24, 223.

2190 J . Org. Chem., Vol. 65, No. 7, 2000
Oh et al.
Ta ble 3. Activa tion P a r a m eter sa for th e Rea ction s of
Y-Ben zylid en em a lon on itr iles w ith X-Ben zyla m in es in
Aceton itr ile
benzylammonium ion in the putative T(. As to the TS
imbalance, the instantaneously localized negative charge
on Câ may facilitate the proton shift to Câ in the TS.
X
Y
∆Hq/kcal mol-1
-∆Sq/cal mol-1 K-1
p-OMe
p-OMe
p-Cl
p-Cl
a
p-Me
p-NO2
p-Me
5.6
5.8
6.1
5.8
32
35
33
38
p-NO2
Calculated by the Eyring equation. The maximum errors
calculated (by the method of Wiberg, K. B. Physical Organic
Chemistry; Wiley: New York, 1964; p 378) are (0.5 kcal mol-1
and (2 eu for ∆Hq and ∆Sq, respectively.
There remains an important result of the negative FY
for the addition of benzylamines to BMN in contrast to
the positive FY for the addition to NS to be explained
consistently with our proposed mechanism: For this
purpose, we should consider relative extent of resonance
delocalization within the reactants, BMN and NS (II and
III), as well as that within the activating groups of the
carbanion intermediates. The resonance delocalization in
+
polarization of the R-carbon to CR should be much
stronger in BMN than in NS.
The relatively lower electron density on CR-Câ due to
strong electron acceptor ability of the (CN)2 group should
lead to smaller primary kinetic isotope effects7 for the
reaction of NS (kH/kD = 2.30-3.08).4
There is almost a linear increase of kH/kD with σY (e.g.,
for X ) p-MeO, kH/kD ) 0.36σY + 2.45, r ) 0.981)
suggesting that a greater degree of bond formation (âx
increases with σY) results in a greater degree of proton
transfer in the TS, I.
The cross-interaction constant, FXY, calculated by sub-
jecting 20 rate data to multiple linear regression using
eq 2a is -0.31. This is also reasonable considering the
values for SN2 reaction (FXY ) - 0.6 to -0.8)5a if we
account for the falloff (=0.47)8 due to an intervening
group, CH2, in benzylamine (leading to FXY = -0.7). The
negative sign of FXY also ensures us an involvement of
the normal N-CR bond formation process in the TS,
excluding any possibilities of T( intermediate formation
for which the sign of FXY has been found to be positive in
the carbonyl addition reactions.9 The smaller magnitude
of FXY found for the addition to BMN than to NS is again
a manifestation of the dissociative nature of the TS for
the addition to BMN, since the magnitude of FXY is a
measure of the degree of bond formation.
II should be stronger than that in III since the two CN
groups provide a stronger π-electron-accepting ability.
The resonance parameters such as R (Swain-Lupton)
and σR (NMR chemical shift) show that two CN groups
are much stronger resonance electron acceptors than an
NO2 group; R (0.30 for (CN)2 and 0.13 for NO2) and σR
(0.36 for (CN)2 and 0.16 for NO2).6 On the other hand,
resonance stabilization of the carbanion in T( by H(NO2)
groups in NS is known to be much stronger than that by
the (CN)2 groups in BMN.2 However, this should apply
to the reactions in aqueous solution, where the terminal
anionic NO2 group in NS is strongly stabilized by hydra-
tion, IV, while stabilization by hydration of the anionic
(CN)2 groups in BMN is small, V. This difference in the
extent of solvation between the two, IV and V, should
be the main reason for the large difference in the strength
of resonance stabilization of the carbanions on the two
activating groups, (CN)2 , H(NO2), in aqueous solution.
In acetonitrile, however, weaker solvation of carbanions
on the two activating groups, H(NO2) in NS and (CN)2
in BMN, may cause to reduce the large difference of
solvating power, and the stronger electron withdrawing
power of the (CN)2 group in BMN may prevail.
The activation parameters, ∆Hq and ∆Sq in Table 3,
are consistent with the concurrent addition of nucleophile
and proton transfer, I.
The bond formation process normally requires small
activation energy, since exclusion repulsion energy is
partially offset by the bond energy of the forming bond.
In the present reactions, a π bond is broken concurrently,
but this is facilitated by the imbalance of the TS in which
reactant resonance stabilization is lost ahead of the bond
formation. The large negative entropy of activation (-32
to -38 eu) is consistent with the four-membered type of
constrained TS structure, I.
In summary, the addition of benzylamine (BA) to
benzylidenemalononitrile (BMN) occurs in a single step
in which the addition of BA to CR of BMN and proton
transfer from BA to Câ of BMN take place concurrently
with a four-membered cyclic TS structure, I. The reaction
center carbon, CR, becomes more positive (FY < 0) on going
from the reactant to transition state due to strong
polarization of the R-carbon, CR+, in the TS by the strong
electron-withdrawing polar effects of the two CN groups.
A measure of the electron-withdrawing power of the
(CN)2 group in the TS can be provided by the R- value,
which has been defined for π-electron acceptor substit-
uents in the reaction with strong conjugation with
electron-rich reaction centers. The R- values for (CN)2
and NO2 are 0.98 and 0.62, respectively;6 these are much
greater than the R values (0.30 and 0.13, respectively),
and the difference between the two (∆R- ) 0.36) is also
very large. Since both the inductive and resonance
electron-withdrawing power of the (CN)2 group are much
stronger than those of the NO2 group (vide supra),
(8) (a) Page, M.; Williams, A. Organic and Bio-organic Mechanisms;
Longman: London, 1997; p 250. (b) Hansch, C.; Hoekman, D.; Gao,
H. Chem. Rev. 1996, 96, 1045.
(9) (a) Lee, I. Bull. Korean Chem. Soc. 1994, 15, 985. (b) Oh, H. K.;
Shin, C. H.; Lee, I. J . Chem. Soc., Perkin Trans. 2 1995, 1169. (c) Lee,
I.; Koh, H. J . New J . Chem. 1996, 20, 131. (d) Koh, H. J .; Kim, O. S.;
Lee, H. W.; Lee, I.. J . Phys. Org. Chem. 1997, 10, 725. (e) Kim, T. H.;
Huh, C.; Lee, B. S.; Lee, I. J . Chem. Soc., Perkin Trans. 2 1995, 2257.

Addition of Benzylamines to Benzylidenemalononitriles
J . Org. Chem., Vol. 65, No. 7, 2000 2191
Ta ble 4. Th e kobs a n d k2 Va lu es for th e Rea ction s of
Y-Ben zylid en em a lon on itr iles w ith p-OMe-Ben zyla m in e
Va r yin g Con cen tr a tion s in Aceton itr ile a t 15.0 °C
The proton transfer in the TS to Câ of BMN is viable
also due to polarization of double bond in the imbalanced
TS. The sign and magnitude of the cross-interaction
constant, FXY, is comparable to those found in the normal
bond formation processes in the SN2 and addition reac-
tions. The normal kinetic isotope effect (kH/kD > 1.0) and
relatively low ∆Hq and large negative ∆Sq values are also
consistent with the mechanism proposed.
Y
[p-OMe-BA]/M kobs/10-3‚s-1 k2/M-1 s-1 corr coeff
p-CH3
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
0.001
0.002
0.003
0.004
0.005
0.006
0.007
0.008
2.78
6.18
9.50
3.07 ( 0.13
2.57 ( 0.08
1.93 ( 0.08
1.87 ( 0.07
1.11 ( 0.04
0.995
0.997
0.995
0.996
0.996
13.7
14.4
17.7
21.1
25.3
2.88
5.31
8.34
Exp er im en ta l Section
H
Ma ter ia ls. Merck GR acetonitrile was used after three
distillations. The Aldrich benzylamine nucleophiles were used
after recrystallization. Aldrich malononitrile and benzalde-
hydes were used.
11.5
13.9
16.1
17.8
21.1
P r ep a r a tion s of Ben zylid en em a lon on itr iles. The benz-
ylidenemalononitriles were prepared by the literature method
of Zabicky.10 Equimolar amounts of benzaldehyde (10 mmol)
and malononitrile (10 mmol) were dissolved in the minimal
amount of 95% ethanol. After the addition of a few drops of
10% potassium hydroxide solution the mixture was set aside
for 1 h. The precipitate was filtered off and purified by repeated
recrystallizations from ethanol (yield > 85%). Melting points,
IR, and 1H and 13C NMR data were found to agree well with
the literature values.2a
Kin etic Mea su r em en t. The reaction was followed spec-
trophotometrically by monitoring the decrease in the concen-
tration of benzylidenemalononitrile, [BMN], at λmax of the
substrate to over 80% completion. The reaction was studied
under pseudo-first-order conditions, [BMN] ) 8 × 10-5 M and
[BA] ) (1.0∼8.0) × 10-3 M at 15.0 ( 0.1 °C. The pseudo-first-
order rate constant, kobs, was determined from the slope of the
plot of ln[BMN] vs time (r > 0.994). Second-order rate
constants, k2, were obtained from the slope of the plot of kobs
vs [BA] (r > 0.995) with more than eight concentrations of
benzylamine, carried out more than three runs, and were
reproducible to within (3%. Typical examples of kobs vs [BA]
data are shown in Table 4.
P r od u ct An a lysis. Benzylidenemalononitrile (0.05 mol)
and benzylamine (0.06 mol) were reacted in acetonitrile at 15.0
°C. After more than 15 half-lives, solvent was removed under
reduced pressure and product was separated by column
chromatography (silica gel, 20% ethyl acetate-n-hexane).
Analytical data of the product gave the following results.
p-CH3C6H5CH(NHCH2C6H4-p-OCH3)CH(CN)2: mp 213-
15 °C, IR(KBr), 3341 (NH, stretch), 3057 (CH, alkene), 2919
(CH, CH3), 2233 (CtN), 1589 (CdC, aromatic); 1H NMR (400
MHz, CDCl3), 1.40 (1H, s, NH), 1.43 (2H, s, benzyl), 2.36 (1H,
d, HC(CN)2, J ) 1.08 Hz), 2.40 (1H, d, CHN, J ) 0.68), 2.45
p-Cl
p-Br
p-NO2
1.93
4.48
7.26
8.66
10.8
12.3
14.1
15.8
2.27
4.46
6.81
8.36
10.0
11.5
13.3
16.2
0.864
1.77
2.85
3.94
5.27
6.80
7.41
8.23
(3H, s, CH3), 3.84 (3H, s, OCH3), 6.69-7.82 (8H, m, aromatic
ring); 13C NMR (100.4 MHz, CDCl3), 158.5, 137.5, 130.8, 130.2,
128.5, 117.7, 113.8, 68.2, 59.2, 31.6, 22.7, 14.1.
Ack n ow led gm en t. The authors acknowledge the
financial support of the Korea Research Foundation
made in the program year of 1998.
(10) Zabicky, J . J . Chem. Soc. 1961, 683.
J O991823D