Electrophilicity parameters for benzylidenemalononitriles

Article abstract of DOI:10.1021/jo0344182

Kinetics of the reactions of stabilized carbanions (derived from nitroethane, diethyl malonate, ethyl cyanoacetate, ethyl acetoacetate, acetyl acetone) with benzylidenemalononitriles have been determined in dimethyl sulfoxide solution at 20 °C. The second-order rate constants are employed to determine the electrophilicity parameters E of the benzylidenemalononitriles according to the correlation equation log k (20 °C) = s(E + N). Comparison with literature data shows that this equation allows the semiquantitative prediction of the reactivities of benzylidenemalononitriles toward a wide variety of nucleophiles, including carbanions, enamines, amines, water, and hydroxide.

Full text of DOI:10.1021/jo0344182

Electr op h ilicity P a r a m eter s for Ben zylid en em a lon on itr iles†
Tadeusz Lemek and Herbert Mayr*
Department Chemie, Ludwig-Maximilians-Universita¨t Mu¨nchen, Butenandtstr. 5-13 (Haus F),
D-81377 Mu¨nchen, Germany
herbert.mayr@cup.uni-muenchen.de
Received April 1, 2003
Kinetics of the reactions of stabilized carbanions (derived from nitroethane, diethyl malonate, ethyl
cyanoacetate, ethyl acetoacetate, acetyl acetone) with benzylidenemalononitriles have been deter-
mined in dimethyl sulfoxide solution at 20 °C. The second-order rate constants are employed to
determine the electrophilicity parameters E of the benzylidenemalononitriles according to the corre-
lation equation log k (20 °C) ) s(E + N). Comparison with literature data shows that this equation
allows the semiquantitative prediction of the reactivities of benzylidenemalononitriles toward a
wide variety of nucleophiles, including carbanions, enamines, amines, water, and hydroxide.
In tr od u ction
ence electrophiles, is directly transferable to other Michael
acceptors. If this were the case, eq 1 could also be
employed for describing the rates of typical Michael
additions. We have examined this question by investigat-
ing the reactions of representative carbanions with
benzylidenemalononitriles.
Numerous kinetic investigations have shown that the
rate constants for the reactions of carbocations and
related electrophiles with neutral nucleophiles can be
described by the following equation:1-4
The reactions of benzylidenemalononitriles with sta-
bilized carbanions have previously been reported to yield
a variety of products.8,9 Though carbanions generally
react with the benzylidenemalononitriles with initial
nucleophilic addition, the primarily formed Michael
adducts often undergo consecutive reactions and cannot
be isolated. The course of transformation of the dicyano-
substituted carbanions formed in the initial addition step
log k (20 °C) ) s(N + E)
(1)
where s ) nucleophile-specific parameter, N ) nucleo-
philicity parameter, and E ) electrophilicity parameter.
Subsequently, we have reported that the reactions of
carbanions with quinone methides, which can be consid-
ered as stabilized benzhydryl cations (diarylcarbenium
ions),5 follow similar relationships.6 Kinetic investigations
of the reactions of benzhydryl cations with stabilized
carbanions allowed us to combine the two sets of kinetic
data, and we have presented a nucleophilicity scale
ranging from methyl-substituted benzenes to stabilized
carbanions (Figure 1).7
(7) Lucius, R.; Loos, R.; Mayr, H. Angew. Chem. 2002, 114, 97-
102; Angew. Chem., Int. Ed. 2002, 41, 91-95. The N and s parameters
for 2a change to N ) 21.54 and s ) 0.62 when rate constants for the
reactions of 2a with further quinone methides are included in the
correlation. See Supporting Information.
(8) For Michael adducts or retro-Michael addition products from
reactions of stabilized carbanions with benzylidenemalononitriles,
see: (a) Goncharenko, M. P.; Sharanin, Yu. A.; Turov, A. V. Zh. Org.
Khim. 1993, 29, 1610-1618; Russ. J . Org. Chem. 1993, 29, 1341-1347.
(b) Michaud, D.; Texier-Boullet, F.; Hamelin, J . Tetrahedron Lett. 1997,
38, 7563-7564. (c) El-Shafei, A. K.; El-Sayed, A. M.; Sultan, A.; Abdel-
Ghany, H. Gazz. Chim. Ital. 1990, 120, 197-201. (d) Ma¸kosza, M.;
Kwast, A. Tetrahedron 1991, 47, 5001-5018. (e) Mart´ın, N.; Quinteiro,
M.; Seoane, C.; Soto, J . L. Liebigs Ann. Chem. 1990, 101-104. (f)
Bogdanowicz-Szwed, K.; Czarny, A. J . Prakt. Chem. 1993, 335, 279-
282. (g) Elagamey, A. G. A.; Abdel-Ghany, A.; Sofan, M. A.; Kandeel,
Z. E.; Elnagdi, M. H. Collect. Czech. Chem. Commun. 1987, 52, 1561-
1565. (h) Sadek, K. U.; Hafez, E. A. A.; Mourad Abo-Elfetouh, E.;
Elnagdi, M. H. Z. Naturforsch. B 1984, 39, 824-828. (i) Negm, A. M.;
Abd El-Aal, F. A. E.-M.; Hafez, E. A.; Elnagdi, M. H.; Mostafa, J . M.
N. Phosphorus, Sulfur, Silicon 1995, 106, 1-7.
(9) For 4H-pyrans from reactions of stabilized carbanions with
benzylidenemalononitriles, see: (a) Mart´ın, N.; Seoane, C.; Soto, J . L.
Tetrahedron 1988, 44, 5861-5868. (b) Zayed, S. E.; Elmaged, E. I. A.;
Metwally, S. A.; Elnagdi, M. H. Collect. Czech. Chem. Commun. 1991,
56, 2175-2182. (c) Assy, M. G.; Hassanien, M. M.; Zaki, S. A. Pol. J .
Chem. 1995, 69, 371-375. (d) Quintela, J . M.; Peinador, C.; Moreira,
M. J . Tetrahedron 1995, 51, 5901-5912. (e) Elnagdi, M. H.; Aal, F. A.
M. A.; Yassin, M. Y. J . Prakt. Chem. 1989, 331, 971-976. (f) Sua´rez,
M.; Molero, D.; Salfran, E.; Mart´ın, N.; Verdecia, Y.; Martinez, R.;
Ochoa, E.; Alba, L.; Quinteiro, M.; Seoane, C. Magn. Reson. Chem.
2001, 39, 105-108. (g) Marco, J . L.; Chinchon, P. M. Heterocycles 1999,
51, 1137-1140. (h) Mart´ın, N.; Pascual, C.; Seoane, C.; Soto, J . L.
Heterocycles 1987, 26, 2811-2816.
The question arose whether the order of nucleophilic
reactivities of carbanions, which has been derived with
respect to carbocations and quinone methides as refer-
† Dedicated to Professor Sema L. Ioffe on the occasion of his 65th
birthday.
* To whom correspondence should be addressed. Fax: (+49) 89-
2180-77717.
(1) Reviews: (a) Mayr, H.; Patz, M. Angew. Chem. 1994, 106, 990-
1010; Angew. Chem., Int. Ed. Engl. 1994, 33, 938-957. (b) Mayr, H.;
Kuhn, O.; Gotta, M. F.; Patz, M. J . Phys. Org. Chem. 1998, 11, 642-
654. (c) Mayr, H.; Patz, M.; Gotta, M. F.; Ofial, A. R. Pure Appl. Chem.
1998, 70, 1993-2000. (d) Mayr, H.; Kempf, B.; Ofial, A. R. Acc. Chem.
Res. 2003, 36, 66-77.
(2) For reactions of carbocations with π-nucleophiles, see: Mayr, H.;
Gotta, M. F.; Bug, T.; Hering, N.; Irrgang, B.; J anker, B.; Kempf, B.;
Loos, R.; Ofial, A. R.; Remennikov, G.; Schimmel, H. J . Am. Chem.
Soc. 2001, 123, 9500-9512.
(3) For reactions of carbocations with hydride donors, see: (a) Funke,
M. A.; Mayr, H. Chem. Eur. J . 1997, 3, 1214-1222. (b) Mayr, H.; Lang,
G.; Ofial, A. R. J . Am. Chem. Soc. 2002, 124, 4076-4083.
(4) For reactions of carbocations with n-nucleophiles, see: Minegishi,
S.; Mayr, H. J . Am. Chem. Soc. 2003, 125, 286-295.
(5) Richard, J . P.; Toteva, M. M.; Crugeiras, J . J . Am. Chem. Soc.
2000, 122, 1664-1674.
(6) Lucius, R.; Mayr, H. Angew. Chem. 2000, 112, 2086-2089;
Angew. Chem., Int. Ed. 2000, 39, 1995-1997.
10.1021/jo0344182 CCC: $25.00 © 2003 American Chemical Society
Published on Web 08/05/2003
6880
J . Org. Chem. 2003, 68, 6880-6886

Electrophilicity of Benzylidenemalononitriles
F IGURE 1. Plot of (log k)/s versus N for the reactions of benzhydryl cations and quinone methides with π-nucleophiles and
carbanions.7
is controlled by many factors, e.g., the nature of the
carbanion, the nature of the aryl group in the arylidene-
malononitrile, as well as the reaction conditions.
2c and 2d (pKa ) 13-14, DMSO)10 also underwent
irreversible second-order reactions with 1a , the most
reactive electrophile of this series.
In contrast, combination of the weak bases 2c,e (pKa
≈ 13, DMSO)10 with the weak electrophiles 1b,c pro-
ceeded with incomplete formation of the Michael adducts
3, which underwent slow consecutive transformations
into 5 and 2f or into 7. An example for the rapid
reversible formation of the initial addition product is
shown in Figure 2. Since 3cc does not absorb at λ ) 441
nm (λmax for 1c), the combination of 1c with 2c results
in a rapid initial decay of the absorbance at λ ) 441 nm
by approximately 35% (< 0.1 s). The consecutive slow
transformation into 5cc (5 min) with an absorption
maximum at λ ) 432 nm then gives rise to a subsequent
increase of the absorbance at λ ) 441 nm as shown in
the insert of Figure 2.
When the reactions of 2c,e with 1b,c were performed
with a high excess of the carbanions 2c,e, pseudo-first-
order kinetics for the first reaction step were observed,
and the pseudo-first-order rate constants kobs were ob-
tained as the slopes of the plots of ln(At - A∞) vs t,
where At is the absorbance of 1 at t and A∞ is the
absorbance of 1 at the time when the equilibrium
between the reactants and the initial adduct 3 is reached.
According to ref 11, the pseudo-first-order rate constants
of reversible reactions represent the sum of forward and
backward reactions (eq 2), and the second-order rate
constants kf can be obtained as the slopes of plots of kobs
versus [2] (Figure 3).
Resu lts a n d Discu ssion
P r od u ct Stu d ies. Three types of products for the
reactions of the carbanions 2a -e with the benzylidene-
malononitriles 1a -c were obtained in this investigation
(Scheme 1):
(1) The ordinary Michael adduct 4a b was produced in
high yield (89%) from the reaction of 1a with 2b in
analogy to the previously reported quantitative formation
of the Michael adduct 4a a from 1a and 2a .8b
(2) In contrast, the malononitrile anion 2f along with
the benzylidenecyanoacetates 5bc and 5cc were obtained
by the reactions of 2c with 1b and 1c, respectively. The
high acidity of malononitrile and the push-pull-stabili-
zation of 5bc and 5cc provide the thermodynamic driving
force for the retro-Michael addition of 6.
(3) In accord with refs 9a,b,d,h, which reported the
formation of the 4H-pyrans 7 in the reactions of 1a ,b with
the acetyl substituted carbanions 2d and 2e, we also
observed the formation of 7be when 1b was combined
with 2e.9a,b We have not been able to isolate the initial
Michael adduct 4be from this combination, even under
mild reaction conditions.
Kin etics. The stopped-flow technique was employed
to determine the kinetics of these reactions. Since the
benzylidenemalononitriles 1a -c show absorption max-
ima at 311, 354, and 441 nm, respectively, and the
products (except 5) do not absorb at these wavelengths,
the progress of the reactions can be monitored photo-
metrically.
kobs ) ([2]kf + kr)
(2)
Second-order kinetics with complete disappearance of
the benzylidenemalononitrile absorptions were observed
for the reactions of 1a -c with the strong bases 2a ,b (pKa
≈ 16, DMSO).10 Analogously, the less basic carbanions
(10) (a) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456-463. (b)
Matthews, W. S.; Bares, J . E.; Bartmess, J . E.; Bordwell, F. G.; Corn-
forth, F. J .; Drucker, G. E.; Margolin, Z.; McCallum, R. J .; McCollum,
G. J .; Vanier, N. R. J . Am. Chem. Soc. 1975, 97, 7006-7014. (c)
Bordwell, F. G.; Fried, H. E. J . Org. Chem. 1981, 46, 4327-4331.
J . Org. Chem, Vol. 68, No. 18, 2003 6881

Lemek and Mayr
F IGURE 3. Pseudo-first-order rate constant kobs for the
reaction of p-dimethylaminobenzylidene-malononitrile 1c with
the carbanion of ethyl cyanoacetate 2c in DMSO at 20 °C as
a function of [2c].
F IGURE 2. Michael addition of 2c to 1c and consecutive
retro-Michael addition with formation of 5cc and 2f monitored
at λ ) 441 nm in DMSO, 20 °C.
SCHEME 2. F or m a tion of 4H-Ch r om en es in
Rea ction s of Dim ed on e An ion w ith
Ben zylid en em a lon on itr iles9c,e,f
SCHEME 1. P r od u cts of Rea ction s of
Ben zylid en em a lon on itr iles 1a -c w ith Ca r ba n ion s
2a -e
TABLE 1. Ra te Con sta n ts for Ad d ition Rea ction s of
Ca r ba n ion s 2a -e w ith Ben zylid en em a lon on itr iles 1a -c
in DMSO a t 20 °C
carb-
pKa
electro- isolated
anion cation (DMSO)a phile product
k2 (M-1 s-1
)
+
2a N(nBu)4
2b K+
16.7
16.4
13.1
14.3
16.7
16.4
13.1
13.3
16.7
16.4
13.1
13.3
1a
1a
1a
1a
1b
1b
1b
1b
1c
1c
1c
1c
4a a b >2 × 107
4a b
(1.15 ( 0.06) × 107
(5.42 ( 0.11) × 106
2c K+
2d K+
7a d c (3.52 ( 0.16) × 106
(5.17 ( 0.23) × 106
+
+
2a N(nBu)4
2b K+
(2.04 ( 0.06) × 106
2c K+
5bc
(8.42 ( 0.11) × 105d
2e K+
7bee (5.96 ( 0.05) × 104d
2a N(nBu)4
(1.85 ( 0.03) × 105
2b K+
(3.34 ( 0.13) × 104
2c K+
5cc
(1.56 ( 0.09) × 104d
2e K+
(1.17 ( 0.07) × 103d
From ref 10. b From ref 8b. c From refs 9d and 9h. Reversible
a
d
CC bond formation. e From refs 9a and 9b.
consumption of 1a -c due to the formation of the 4H-
chromene 9 (Scheme 2),9c,e,f cyclization with irreversible
formation of 9 occurs in a slow consecutive reaction, and
we have not been able to find conditions under which
second-order kinetics was obeyed. The order with respect
to carbanion was generally greater than 1, usually close
to 2, and addition of dimedone as a proton-source did not
simplify the kinetics as in the analogous reactions of 2g
with quinone methides.6 For that reason, kinetic data for
the reactions of 2g are not included in Table 1.
Deter m in a tion of th e Electr op h ilicity P a r a m -
eter s. Equation 1 was used to calculate the E parameters
of 1a -c from the rate constants given in Table 1 and
the previously reported N and s parameters of the
carbanions 2a -e.7 A least-squares fit of calculated and
experimental rate constants (minimization of ∆2 ) Σ(log
k - s(N + E))2 with What’sBest!’s nonlinear solver) gave
the E parameters of the benzylidenemalononitriles
Reversibility of the nucleophilic attack was a problem
in all reactions of the weakly basic anion of dimedone
(2g). Though the reactions of 2g proceed with complete
(11) Schmid, R.; Sapunov, V. N. Non-Formal Kinetics; Ebel, H. F.,
Ed.; Monographs in Modern Chemistry 14; Verlag Chemie: Weinheim,
1982; pp 20-22.
6882 J . Org. Chem., Vol. 68, No. 18, 2003

Electrophilicity of Benzylidenemalononitriles
F IGURE 4. Rate constants for the reactions of carbanions 2a -e with benzylidenemalononitriles 1a -c compared with the
reactivities toward reference electrophiles.7 The rate constants for the reactions of 1a -c with 2a -e are not used for the construction
of the correlation lines.
1a -c, which are close to the arithmetic means of E
calculated from k of the individual experiments. A com-
parison of the electrophilicities of benzylidenemalononi-
triles 1a -c with the electrophilicities of benzhydryl cat-
ions and quinone methides is given in Figure 4. The good
fit demonstrates that the nucleophilic reactivity order of
carbanions in DMSO as derived from the reactions with
benzhydryl cations and quinone methides, also holds for
the reactions with typical Michael acceptors.
Discu ssion
A three-parameter equation, like eq 1, cannot be
expected to reproduce and predict rate constants for
electrophile nucleophile combinations of large structural
variety with high precision. However, the satisfactory fit
of the rate constants determined in this investigation to
the previously published correlation lines7 in Figure 4
shows the suitability of this approach for a semiquanti-
tative treatment of organic reactivity. From the E
parameters given in Figure 4, one can now derive that
the noncharged Michael acceptors 1a -c have electrophi-
licities comparable to those of amino-substituted ben-
zhydrylium ions or triarylallyl cations,1d,12 allylpalladium
complexes,1d,13 or tricarbonylcycloheptadienylium cat-
F IGURE 5. Comparison of the electrophilicity parameters E
of charged1d and noncharged electrophiles.
ions,1d,14 i.e., Figure 5 illustrates the electrophilicity range
where highly stabilized carbocations and strongly electron-
The malonodinitrile anion 2f was found to possess
deficient noncharged π-electrophiles overlap.
similar nucleophilicity parameters in DMSO and in
To examine the suitability of the electrophilicity pa-
water,15 in accord with Bernasconi’s report that this
rameters determined in this work for generally predicting
carbanion reacts only four times faster in 50% aqueous
the electrophilic reactivities of the benzylidenemalono-
nitriles 1a -c, we have calculated the rate constants of
their reactions with other types of nucleophiles by eq 1
and compared these numbers with experimental values
(Table 2).
DMSO than in pure water.16 The rate constant calculated
for the reaction of 2f with 1a is 1.3 times larger than
that measured by Bernasconi.16 In view of the ap-
proximate character of the N and s paramters for CN-/
(14) Mayr, H.; Mu¨ller, K.-H.; Rau, D. Angew. Chem. 1993, 105,
1732-1734; Angew. Chem., Int. Ed. Engl. 1993, 32, 1630-1632.
(15) Bug, T.; Mayr, H. Unpublished results.
(16) Bernasconi, C. F.; Zitomer, J . L.; Fox, J . P.; Howard, K. A. J .
Org. Chem. 1984, 49, 482-486.
(12) (a) Mayr, H.; Fichtner, C.; Ofial, A. R. J . Chem. Soc., Perkin
Trans. 2 2002, 1435-1440.
(13) Kuhn, O.; Mayr, H. Angew. Chem. 1999, 111, 356-358; Angew.
Chem., Int. Ed. 1999, 38, 343-346.
J . Org. Chem, Vol. 68, No. 18, 2003 6883

Lemek and Mayr
TABLE 2. Com p a r ison of Exp er im en ta l Secon d -Or d er Ra te Con sta n ts kexp a n d kca lc Ca lcu la ted (fr om eq 1) for
Rea ction s of Ben zylid en em a lon on itr iles 1a -c w ith Va r iou s Nu cleop h iles
electrophile
(E parameter)
nucleophile
N
s
k20°Ccalc (M-1 s-1
)
kexp (M-1 s-1) (conditions)
19.58a
19.36c
9.19d
0.54a
0.67c
0.60d
0.44f
0.38h
0.67j
0.89l
0.61n
0.82°
0.38h
0.61n
0.38h
3.1 × 105 (H2O)
2.3 × 105 (20 °C, H2O)b
-
1a (-9.42)
CH(CN)2
4.6 × 106 (DMSO)
0.73 (H2O)
9.5 × 105 (20 °C, H2O/DMSO 1/1 v/v)b
0.50 (25 °C, H2O/EtOH 4/1 v/v)e
2.6 × 102 (20 °C, MeCN)g
CN-
benzylamine
piperidine
morpholine
H2O
14.7f
2.3 × 102 (MeCN)
20.9h
2.30 × 104 (H2O-DMSO 1/1 v/v)
1.1 × 105 (DMSO)
1.8 × 105 (20 °C, H2O/DMSO 1/1 v/v)i
1.6 × 105 (20 °C, DMSO/H2O 7/1 v/v)k
1.2 × 10-4 s-1 (20 °C, H2O)m
16.96j
5.11l
1.5 × 10-4 s-1 (H2O)
4.4 (H2O)
OH-/H2O
10
10.47n
10.04°
20.9h
1.3 × 102 (20 °C, H2O)m
3.2 (CH2Cl2)
1.2 × 10-2 (20 °C, CH2Cl2)p
1b (-10.8)
piperidine
OH-/H2O
piperidine
6. 9 × 103 (H2O-DMSO 1/1 v/v)
9.8 × 104 (20 °C, H2O/DMSO 1/1 v/v)k
2.1 × 101 (20 °C, H2O)m
10.47n
20.9h
0.63 (H2O)
1c (-13.3)
7.7 × 102 (H2O-DMSO 1/1 v/v)
5.6 × 103 (20 °C, H2O/DMSO 1/1 v/v)k
a
b
d
N and s for CH(CN)2-/H2O (from ref 15). From ref 16. c N and s for CH(CN)2-/DMSO (from ref 7). For approximate values for
g
h
CN-/H2O see ref 4. e From ref 17. f N and s for n-PrNH2 in MeCN (from ref 18). From ref 19. N and s in 50% aqueous DMSO (see
Supporting Information); the plot of log k vs E as shown in Figure S7 has an unusually low correlation coefficient that we do not recommend
i
j
k
l
to use these numbers for further calculations. From ref 20. N and s for morpholine in DMSO (from ref 4). From ref 21. N and s refers
to first-order rate constants (see ref 4). From ref 22. From ref 4. o For N and s of 10 in CH2Cl2, see ref 23. Extrapolated value, see
m
n
p
Supporting Information.
SCHEME 3. Rever sible Step w ise [2 + 2] Cycloa d d ition of Mor p h olin oisobu tylen e 10 w ith
Ben zylid en em a lon on itr ile 1a
TABLE 3. Ra te a n d Equ ilibr iu m Con sta n ts for [2 + 2]
H2O,4 the excellent agreement between calculated and
Cycloa d d ition of Ben zylid en em a lon on itr ile 1a w ith
experimental numbers for the reactions of 1a with CN-
N-Mor p h olin oisobu ten e 10 (CH2Cl2)
in water has to be considered as accidental. Entry 3 of
T (°C) kf (M-1 s-1
)
kr (s-1
)
Ka (M-1
)
Kb (M-1
)
Table 2 shows that the calculated rate constant for the
reaction of n-propylamine with 1a in acetonitrile is
almost identical to the published experimental value19
for the reaction of benzylamine with 1a in the same
solvent.
Nucleophilicity parameters N and s for piperidine and
morpholine have previously only been established for the
solvent DMSO.4 To compare predictions by eq 1 with the
published rate constants for the reactions of piperidine
with 1a -c in 50% aqueous DMSO,21 we have now
determined N ) 20.9 and s ) 0.38 for piperidine in 50%
aqueous DMSO as described in the Supporting Informa-
tion. With these numbers, we can now calculate the rate
constants for the reactions of piperidine with 1a -c in
50% aqueous DMSO, and as shown in Table 2, the three
calculated rate constants are 7-13 times smaller than
those measured by Bernasconi.20,21
-20.0 2.81 × 10-3
-41.0 1.19 × 10-3 (7.3 × 10-6
-50.0 7.10 × 10-4 (3.5 × 10-6
2.0 × 10-4
14
46
(163)
(205)
19 (from two runs)
48 (from three runs)
103 (from three runs)
-30.5 2.03 × 10-3
4.4 × 10-5
)
)
a
b
Calculated from K ) kf/ kr. Calculated from K ) (A0
-
A∞)/(A∞[10]) at λmax
.
OH- in water was reported to react 30 times faster with
1a and 1b than predicted by eq 1.
Because rate constants for the reactions of benzylidene-
malononitriles with π-nucleophiles of known N and s
parameters have not been available in the literature, we
have studied the kinetics of the reaction of 1a with the
enamine 10 in this work (Scheme 3). In accord with
literature reports, the previously characterized cyclobu-
tane 12 is formed,24 presumably via the dipolar interme-
diate 11.
The incomplete disappearance of the absorbance of 1a
indicated the existence of an equilibrium, and when 10
was used in high excess, it was possible to derive kf and
kr from a plot of the pseudo-first-order rate constant kobs
versus [10] in analogy to the procedure described above
(cf. Figure 3). Table 3 compares the equilibrium constants
derived as the ratio of forward over backward rate
constants with those derived from the UV-spectroscopi-
It must again be considered as a fortuity that the rate
constant calculated for the reaction of morpholine with
1a in DMSO differs by only a factor of 1.4 from the one
measured in 70% DMSO/30% H2O. While the first-order
rate constant calculated for the reaction of 1a with water
is exactly matching the one measured by Bernasconi,22
(17) Pritchard, R. B.; Lough, C. E.; Reesor, J . B.; Holmes, H. L.;
Currie, D. J . Can. J . Chem. 1967, 45, 775-777.
(18) Ruppert, T. Diplomarbeit; Ludwig-Maximilians-Universita¨t
Mu¨nchen, 1998.
(19) Varghese, B.; Kothari, S.; Banerji, K. K. Int. J . Chem. Kinet.
1999, 31, 245-252.
(22) Bernasconi, C. F.; Fox, J . P.; Kanavarioti, A.; Panda, M. J . Am.
Chem. Soc. 1986, 108, 2372-2381.
(20) Bernasconi, C. F.; Fox, J . P.; Fornarini, S. J . Am. Chem. Soc.
1980, 102, 2810-2816.
(23) Kempf, B.; Hampel, N.; Ofial, A. R.; Mayr, H. Chem. Eur. J .
2003, 9, 2209-2218.
(21) Bernasconi, C. F.; Killion, R. B. J . Org. Chem. 1989, 54, 2878-
2885.
(24) Penades, S.; Kisch, H.; Tortschanoff, K.; Margaretha, P.;
Polansky, O. E. Monatsh. Chem. 1973, 104, 447-456.
6884 J . Org. Chem., Vol. 68, No. 18, 2003

Electrophilicity of Benzylidenemalononitriles
was used. The benzylidenemalononitriles 1b -c were pre-
pared according to literature procedures32 and recrystallized
from EtOH. Potassium or tetra-n-butylammonium salts of
carbanions 2b-f were prepared according to published pro-
cedures.33
cally determined concentrations of 1a . Though the two
methods give somewhat different results, it is obvious
that the cyclobutane is favored at low temperature,
whereas dissociation is preferred at higher temperature.
From a van’t Hoff plot, ∆rH° ) -39.2 kJ mol-1 and ∆rS°
) -130 J mol-1 K-1 (K from concentrations) can be
derived.
An Eyring plot for the reaction of 1a with 10 yields
∆Hq ) 19.8 kJ mol-1 and ∆Sq ) -214 J mol-1 K-1, in
agreement with Huisgen’s reports that stepwise [2 + 2]
cycloadditions via zwitterionic intermediates generally
proceed with highly negative entropies of activation.25
We can now use the Eyring parameters to calculate
the combination rate constant for 1a + 10 at 20 °C
(0.0124 M-1 s-1) and compare it with the value derived
from eq 1 (3.2 M-1 s-1). Though the two numbers differ
by a factor of 250, it is remarkable that eq 1 is even able
to predict the correct order of magnitude for stepwise [2
+ 2] cycloadditions. A better agreement cannot be
expected because of the Coulomb interactions in the
intermediate 1,4-dipole, which strongly depend on solvent
polarity.26
P r od u ct Stu d ies. Rea ction of 1a w ith 2b. Diethyl
malonate 2b (305 µL, 2.00 mmol) was added to a solution of
freshly sublimated tBuOK (225 mg, 2.00 mmol) in DMSO (1
mL) at room temperature. After 2 min of stirring, a solution
of 1a (154 mg, 1.00 mmol) in DMSO (1 mL) was added. After
10 min, trifluoroacetic acid (200 µL, 2.60 mmol) was added,
and the mixture was poured into water (50 mL). A white solid
precipitated, which was filtered, washed with water, and dried
in the air. The crude product was dissolved in 2 mL of
methanol and crystallized to give 280 mg (89%) of 2-(2,2-
dicyano-1-phenyl-ethyl)-malonic acid diethyl ester (4a b): mp
79-79.5 °C; 1H NMR (CDCl3) δ 0.93 (t, J ) 7.1, 3 H, CH2CH3),
1.28 (t, J ) 7.1, 3 H, CH2CH3), 3.89 (dd, J ) 11.4 J ) 5.0, 1 H,
PhCH), 3.92 (q, J ) 7.1, 2 H, CH2CH3), 4.07 (d, J ) 11.4, 1 H,
CH(CN)2), 4.27 (q, J ) 7.1, 2 H, CH2CH3), 4.88 (d, J ) 5.0, 1
H, CH(CO2Et)2), 7.38 (s, 5 H, Ph); 13C NMR (CDCl3) δ 13.5
(CH3), 13.8 (CH3), 27.6 (CH(CN)2), 44.8 (PhCH), 53.4 (CH(CO2-
Et)2), 62.1 (CH2), 62.8 (CH2), 111.2 (CN), 111.2 (CN), 128.5 (Cm
Ph, tentatively), 129.1 (Co Ph, tentatively), 129.6 (Cp Ph), 133.4
(Ci Ph), 165.8 (CO2Et), 167.5 (CO2Et). Anal. Calcd for
C
17H18N2O4 (314.34): H 5.77 C 64.96 N 8.91. Found: H 5.78
We therefore conclude that the E parameters of the
benzylidenemalononitriles 1a -c derived in this work are
a useful tool for predicting the reactivities of these
electrophiles. It should be noted, however, that this
statement cannot be generalized for all types of Michael
acceptors. The relative reactivities of different nucleo-
philes toward tetracyanoethylene,27 for example, are not
properly reproduced by eq 1, probably because of desta-
bilization of the zwitterionic intermediates by the inverse
anomeric effect of two geminal cyano groups, which
strongly depends on the nature of the additional substit-
uents.28 Despite of this limitation, the results of this
investigation suggest to systematically analyze the po-
tential of eq 1 for predicting the rates of nucleophilic
additions to electron-deficient olefins.29 In view of the
previously reported relationship between eq 1 and Ritch-
ie’s N+ relationship4,30 and the finding by Hoz31 that the
rate constants for nucleophilic additions to acceptor-
substituted 9-methylenefluorenes correlate with Ritchie’s
N+ values, one may expect that many more types of
electrophile nucleophile combinations can be described
by the linear free energy relationship (eq 1).
C 65.15 N 8.98.
Rea ction of 1b w ith 2c. Ethyl cyanoacetate (340 mg, 3.00
mmol) was added to a vigorously stirred solution of freshly
sublimated tBuOK (337 mg, 3.00 mmol) in DMSO (1 mL).
After 1 min of stirring, a solution of 1b (184 mg, 1.00
mmol) in DMSO (2 mL) was added. After 15 min, trifluoro-
acetic acid (250 µL, 3.25 mmol) was added, and the mixture
was poured into water (100 mL). The product was extracted
with AcOEt (3 × 20 mL), and the organic layer was separated,
washed with brine, and dried with CaCl2. The solvent was
removed, and the crude product (a yellow oil, 500 mg) was
dried in vacuo. Crystallization from MeOH and recrystalliza-
tion from hexane/AcOEt yielded 100 mg (43%) of 2-cyano-3-
(4-methoxy-phenyl)-acrylic acid ethyl ester 5bc (the isolation
of the product was not quantitative): mp 80-81 °C (lit.34
82-83 °C). The 1H NMR spectrum was consistent with that
reported in ref 34.
Rea ction of 1c w ith 2c. Ethyl cyanoacetate (340 mg, 3.00
mmol) was added to a vigorously stirred suspension of K2CO3
(690 mg, 5.00 mmol) in DMSO (2 mL). After addition of 1c
(99 mg, 0.500 mmol) the mixture was stirred for 45 min at
room temperature and diluted with 20 mL of AcOEt and 10
mL of aqueous HCl (1 M). The organic layer was separated,
washed with water and brine, dried with Na2SO4, and purified
by column chromatography on silica gel 60 F254 (eluent hex-
ane/AcOEt). Recrystallization from hexane/AcOEt yielded 110
mg (90%) of 2-cyano-3-(4-(dimethylamino)-phenyl)acrylic acid
Exp er im en ta l Section
ethyl ester (5cc): mp 125-126 °C (lit.34 127-128 °C). The H
NMR spectrum of 5cc was consistent with that reported
in ref 34. The UV spectrum shows λmax ) 432 nm in DMSO at
20 °C.
Rea ction of 1b w ith 2e. Triethylamine (50 µL, 0.36 mmol)
was added to a solution of 1b (92 mg, 0.50 mmol) and acetyl
acetone (100 mg, 1 mmol) in DMSO (2 mL). The mixture was
stirred for 1h at room temperature and diluted with AcOEt
(20 mL) and aqueous HCl (5 mL, 1 M). The organic layer was
separated, washed with water and brine, dried with Na2SO4,
1
In str u m en ts a n d Ma ter ia ls. 1H and 13C NMR spectra
were recorded on 300 and 75.5 MHz instruments, respectively.
Chemical shifts are expressed in ppm and refer to Me4Si (δH
) 0.00 ppm), coupling constants are in Hz, and melting points
are uncorrected. The yields refer to isolated products without
optimization of the procedures. DMSO 99.5% (H2O < 0.01%)
(25) Steiner, G.; Huisgen, R. Tetrahedron Lett. 1973, 14, 3769-3772.
(26) Steiner, G.; Huisgen, R. J . Am. Chem. Soc. 1973, 95, 5056-
5058.
(27) (a) Huisgen, R.; Steiner, G. Tetrahedron Lett. 1973, 39, 3763-
3768. (b) Huisgen, R.; Schug, R.; Steiner, G. Angew. Chem. 1974, 86,
47-48; Angew. Chem., Int. Ed. Engl. 1974, 13, 80. (c) Huisgen, R. Acc.
Chem. Res. 1977, 10, 199-206.
(32) Corson, B. B.; Stoughton, R. W. J . Am. Chem. Soc. 1928, 50,
2825-2837.
(28) Beckhaus, H.-D.; Dogan, B.; Pakusch, J .; Verevkin, S.; Ru¨cha-
rdt, C. Chem. Ber. 1990, 123, 2153-2159.
(33) (a) Arnett, E. M.; Maroldo, S. G.; Schilling, S. L.; Harrelson, A.
J . Am. Chem. Soc. 1984, 106, 6759-6767. (b) Reetz, M. T.; Hu¨tte, S.;
Goddard, R. Eur. J . Org. Chem. 1999, 2475-2478. (c) Reetz, M. T.;
Hu¨tte, S.; Goddard, R. Z. Naturforsch. B 1995, 50, 415-422. (d) Reetz,
M. T.; Hu¨tte, S.; Goddard, R. J . Am. Chem. Soc. 1993, 115, 9339-
9340.
(29) For an authoritative review, see: Bernasconi, C. F. Tetrahedron
1989, 45, 4017-4090.
(30) (a) Ritchie, C. D. Acc. Chem. Res. 1972, 5, 348-354. (b) Ritchie,
C. D. Can. J . Chem. 1986, 64, 2239-2250.
(31) Hoz, S.; Speizman, D. J . Org. Chem. 1983, 48, 2904-2910.
(34) Shen, Y.; Yang, B. Synth. Commun. 1989, 19, 3069-3075.
J . Org. Chem, Vol. 68, No. 18, 2003 6885

Lemek and Mayr
and purified by column chromatography on silica gel 60 F254
with hexane/AcOEt solvent system. Recrystallization from
hexane/CHCl3 yielded 125 mg (88%) of 5-acetyl-2-amino-4-(4-
methoxyphenyl)-6-methyl-4H-pyran-3-carbonitrile 7be: mp
Deutsche Forschungsgemeinschaft and the Fonds der
Chemischen Industrie for financial support.
Su p p or tin g In for m a tion Ava ila ble: Details of the ki-
netic experiments and NMR spectra of 2,6-di-tert-butyl-4-
(piperidin-1-yl-(4-tolyl)methyl)phenol. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
1
158-159 °C (lit.9a,b 157-159 °C). The H NMR spectrum was
in agreement with the spectrum reported in ref 9a.
Kin etic Mea su r em en ts. The kinetic measurements were
performed as previously reported.35
J O0344182
Ack n ow led gm en t. We thank the Alexander von
Humboldt Foundation for a fellowship to T.L. and the
(35) Dilman, A. D.; Ioffe, S. L.; Mayr, H. J . Org. Chem. 2001, 66,
3196-3200.
6886 J . Org. Chem., Vol. 68, No. 18, 2003