Organocatalyzed and uncatalyzed C=C/C=C and C=C/C=N exchange processes between knoevenagel and imine compounds in dynamic covalent chemistry

Article abstract of DOI:10.1002/hlca.201400187

Molecular diversity generation through reversible component exchange has acquired great importance in the last decade with the development of dynamic covalent chemistry. We explore here the recombination of components linked by C=C and C=N bonds through reversible double-bond formation, and cleavage in C=C/C=C and C=C/C=N exchange processes. The reversibility of the Knoevenagel reaction has been explored, and C=C/C=C C/C exchanges have been achieved among different benzylidenes, under organocatalysis by secondary amines such as L-proline. The substituents of these benzylidenes were shown to play a very important role in the kinetics of the exchange reactions. L-Proline is also used to catalyze the reversible C=C/C=C exchange between Knoevenagel derivatives of barbituric acid and malononitrile. Finally, the interconversion between Knoevenagel derivatives of dimethylbarbituric acid and imines (C=C/C=N exchange) has been studied and was found to occur rapidly in the absence of catalyst. The results of this study pave the way for the extension of dynamic combinatorial chemistry based on C=C/C=C and C=C/C=N exchange systems.

Full text of DOI:10.1002/hlca.201400187

Helvetica Chimica Acta – Vol. 97 (2014)
1219
Organocatalyzed and Uncatalyzed C¼C/C¼C and C¼C/C¼N Exchange
Processes between Knoevenagel and Imine Compounds in Dynamic Covalent
Chemistry
by Sirinan Kulchat, Kamel Meguellati, and Jean-Marie Lehn*
´
´
´
Laboratoire de Chimie Supramoleculaire, Institut de Science et dꢀIngenierie Supramoleculaires (ISIS),
´
´
Universite de Strasbourg, 8 allee Gaspard Monge, BP 70028, F-67000 Strasbourg Cedex
(phone: þ 33368855145; fax: þ 33368855140; e-mail: lehn@unistra.fr)
Molecular diversity generation through reversible component exchange has acquired great
importance in the last decade with the development of dynamic covalent chemistry. We explore here
the recombination of components linked by C¼C and C¼N bonds through reversible double-bond
formation, and cleavage in C¼C/C¼C and C¼C/C¼N exchange processes. The reversibility of the
Knoevenagel reaction has been explored, and C¼C/C¼C C/C exchanges have been achieved among
different benzylidenes, under organocatalysis by secondary amines such as l-proline. The substituents of
these benzylidenes were shown to play a very important role in the kinetics of the exchange reactions. l-
Proline is also used to catalyze the reversible C¼C/C¼C exchange between Knoevenagel derivatives of
barbituric acid and malononitrile. Finally, the interconversion between Knoevenagel derivatives of
dimethylbarbituric acid and imines (C¼C/C¼N exchange) has been studied and was found to occur
rapidly in the absence of catalyst. The results of this study pave the way for the extension of dynamic
combinatorial chemistry based on C¼C/C¼C and C¼C/C¼N exchange systems.
1. Introduction. – ꢁConstitutional Dynamic Chemistryꢀ (CDC) [1] involves the
dynamic recombination of molecular components linked through either non-covalent
interactions or reversible covalent bonds. In recent years, the latter, dynamic covalent
chemistry (DCC), has emerged as a powerful approach to create dynamic libraries of
compounds of increasing structural diversity. The screening of these ꢁDynamic
Covalent Librariesꢀ (DCLs) provides new perspectives in drug design [2] and material
sciences [3] (for dynamic materials, see [4]). Representative examples of reversible
reactions compatible with the DCC concept include amine/carbonyl condensations
[5][6], transesterifications [7], disulfide exchange [8], peptide exchange [9], boronic
ester formation [10], olefin metathesis [11], and DielsꢀAlder condensation [12].
However, there is a constant need for new reversible reactions to create DCLs of
increased structural and chemical diversity, and complexity.
Organocatalysis [13] by primary and secondary amines via enamine and iminium
intermediates has been actively pursued in recent years to facilitate carbonyl
condensation processes such as Knoevenagel and aldolization reactions [14]. The
Knoevenagel reaction consists in the condensation of an activated hydrogen compound
such as b-keto esters and malonates with aldehydes or ketones [15a] (for an excellent
review, see [15b]) in the presence of primary or secondary amines, resulting in the
formation of a C¼C bond. Most frequently employed secondary amine catalysts include
ꢂ 2014 Verlag Helvetica Chimica Acta AG, Zꢃrich

1220
Helvetica Chimica Acta – Vol. 97 (2014)
piperidine, sarcosine, and proline and its derivatives, the latter being extensively used
for asymmetric organocatalysis [13].
In the present study, we have investigated the DCC processes involving reversibility
and component exchange in Knoevenagel reactions through cleavage and reformation
of C¼C bonds between the condensation compounds of barbituric acid B or
malononitrile M, and three benzaldehydes 1a – 1c (Scheme 1). We also explored the
ability of selected secondary amines to catalyze these reactions, in particular using l-
proline as the catalyst. Such processes amount to dynamic formation and cleavage of
C,C double bonds, C¼C, and allow for C¼C/C¼C exchange of either moiety. The
exchange of components has been reported in Michael reactions on benzylidene
compounds [16].
Scheme 1. Benzaldehyde Derivatives 1a – 1c and Their Knoevenagel Condensation Compounds with
Barbituric Acid (B), 2a – 2c, and Malononitrile (M), 3a – 3c
We have reported earlier Knoevenagel/imine (C¼C/C¼N) cross-exchange processes
using l-proline as an organocatalyst [17] in pure (D6)DMSO. As an extension of this
previous work, we also describe here the C¼C/C¼N cross-exchange with Knoevenagel
compounds of dimethyl-barbiturate, achieving fast recombination in the absence of
catalyst in pure CDCl3.
2. Results and Discussion. – 2.1. Synthesis and Hydrolysis of the Benzylidene
Compounds 2a – 2c and 3a – 3c (Scheme 1). A series of benzylidene derivatives, 2a – 2c
and 3a – 3c, were synthesized via Knoevenagel condensation between, respectively,
barbituric acid B and malononitrile M, and 4-(dimethylamino)benzaldehyde (1a),
benzaldehyde (1b), 4-nitrobenzaldehyde (1c). The reversible formation and hydrolysis
processes of these compounds are illustrated in Scheme 1. They were followed by
1H-NMR spectroscopy (see Exper. Part) in the presence of a secondary amine (l-
proline, l-proline methyl ester, piperidine). Such bases, especially l-proline, accel-
erated both the rates of formation and of hydrolysis of the compounds. The
corresponding kinetic (half-life, t1/2) and thermodynamic equilibrium parameters
(equilibrium constants K and corresponding free energies DG8) are compiled in
Table 1.

Helvetica Chimica Acta – Vol. 97 (2014)
1221
Table 1. Kinetic and Equilibrium Thermodynamic Parameters for Knoevenagel Condensation and
Hydrolysis Processesa). Half-life, t1/2; þ , reaction too fast; ꢀ , reaction too slow to determine the
parameter, F ¼ formation, H ¼ hydrolysis, K ¼ equilibrium constant and DG8 ¼ ꢀRT ln(K).
2a
2b
2c
3a
3b
3c
Catalystb)
t1/2 [h] F
t1/2 [h] H
%c) F
%c) H
K
B
4
15
77
77
Pr
B
20
Pr
2
0.2
59
53
B
4.5
Pr
2.1
B
ꢀ
ꢀ
ꢀ
Pr
B
Pr
B
þ
Pr
þ
1
3
þ
þ
0.4 20.5
þ
þ
þ
þ
þ
3
0.7
82
78
57
55
62
61
61
59
95
95
95
91
86
40.4
91
86
40.4
94 95
89
15.2
88
15.2
0.58
1.01 0.12 0.14 0.17 0.16
ꢀ
15.2
DG0
1.35 ꢀ 0.02 5.25 4.87 4.39 4.54
ꢀ
ꢀ 6.74 ꢀ 6.74 ꢀ 6.74 ꢀ 9.16 ꢀ 9.16
[kJ molꢀ1
]
a) Experiments were performed at 608 in (D6)DMSO/D2O 99 : 1 with 10 mol-% base catalyst. b) B,
d
blank; Pr, l-proline. c) %, Percentage of constituents at equilibrium. ) DG8 in kJ molꢀ1
.
The equilibrium constants were obtained from the amounts of reactants and
products according to Eqn. 1:
K ¼ [2]/[1][B] or K ¼ [3]/[1][M]
(1)
The electrophilic behavior of the benzylidene derivatives 2 and 3 is strongly
dependent on the nature of the probe-head (i.e., B or M). Most reactive are the
benzylidene derivatives whose probe-head is the strongest electron-withdrawing group
and has the lowest pKa. Thus, diethyl benzylidenemalonates (pKa (diethyl malonate) ¼
16.4) proved to be less reactive than the benzylidene-malononitrile (pKa
(malononitrile) ¼ 11.1) and much less reactive than the barbituric acid analogs (pKa
(barbituric acid) ¼ 4.0), consistent with a recent study [16]. Because of the very low
reactivity of diethyl benzylidenemalonates, we decided to focus our study on the
barbiturate and malononitrile derivatives only. The stability of the benzylidene
derivatives toward hydrolysis also proved dependent on the nature of the para-
substituent on the benzaldehyde component. The compounds bearing electron-
donating substituents such as dimethylamino (Me2N), i.e., 2a and 3a, were less
reactive than those bearing an electron-attracting substituent (NO2), 2c and 3c.
The effects of the addition of a base during the synthesis (formation) or hydrolysis
of each benzylidene derivative were also examined. The reactions between 1 equiv. of
aldehyde 1a – 1c and 1 equiv. of either barbituric acid B or malononitrile M (12.8 mm
each) were carried out in (D6)DMSO/D2O 99 :1 and at 608, in the presence and in the
absence of base as a potential catalyst (l-proline, l-proline methyl ester, or piperidine),
and the formation of the corresponding conjugated benzylidene derivative was
monitored by 1H-NMR. The reverse (hydrolysis) reactions were performed in parallel,
and the hydrolysis of each separately prepared conjugated benzylidene derivative
(12.8 mm in (D6)DMSO/D2O 99 :1 at 608, in the presence and in the absence of base)
was monitored by 1H-NMR. After 36 h, the same distribution of all reaction components
was observed for both forward (condensation) and backward (hydrolysis) processes,
thus evidencing that the reactions were reversible and under thermodynamic control.

1222
Helvetica Chimica Acta – Vol. 97 (2014)
The effects of the addition of base on the kinetics and thermodynamics of the
reaction were also investigated. Generally, all three bases tested (see Fig. 1) were found
to accelerate both the synthesis and, conversely, the hydrolysis of the conjugated
benzylidene compound. Of the three bases tested, l-proline proved the most efficient
catalyst, followed by l-proline methyl ester. Piperidine came last despite its higher
basicity, possibly because of extensive protonation. These results indicate the likely
participation of l-proline carboxylic acid in the catalytic mechanism [18]. The most
pronounced effects were observed for the formation and hydrolysis of compound 2b
with a Me2N substituent (Fig. 1) for which l-proline accelerated the reaction tenfold
(t1/2 reduced from 20 to 2 h) when compared to the reaction carried out in the absence
of base. It is also noteworthy that the catalytic effect of l-proline was lower for
benzylidenes substituted with an electron-withdrawing group. In fact, such compounds
hydrolyze quickly even in the absence of base, thus catalyst addition has little effect.
2.2. Study of the Knoevenagel Exchange Reaction. The reversibility of the
Knoevenagel reaction involves hydrolytic cleavage of the C¼C bond by a retro-
Knoevenagel process. It can be ascertained by the addition of another aldehyde
component to produce a new benzylidene conjugate with C¼C bond interchange
(Scheme 2), via a sequential hydrolysisꢀcondensation mechanism. The capacity of
various benzylidene derivatives to undergo such exchange reactions in the presence of
various aldehydes was investigated, in the absence or in the presence of different
secondary amines, l-proline, l-proline methyl ester or piperidine as potential catalysts,
for all possible combinations of one benzylidene derivative (i.e., 2a – 2c and 3a – 3c)
with one aldehyde (i.e., 1a – 1c; Scheme 2). Each benzylidene compound (12.8 mm) was
solubilized in (D6)DMSO/D2O 99 :1, heated at 608 together with 1 equiv. of a
differently substituted aldehyde, and in the absence or in the presence of one of the
three bases. For each reaction, the percentage of each component of the reaction
mixture (aldehydes and benzylidene derivatives) as a function of time, as well as the
initial rates of disappearance of the starting benzylidene compound were determined
1
by H-NMR spectroscopy. The reactions were also carried out at 258 and 408, but the
!
Fig. 1. Formation (left) and hydrolysis (right) of compound 2b catalyzed by secondary amines. , No
~
&
*
base added; , piperidine; , l-proline methyl ester; , l-proline. Reactions were carried out in
triplicate, and the standard deviation was < 5%. The curves were drawn through the experimental points
for clarity.

Helvetica Chimica Acta – Vol. 97 (2014)
1223
exchange became very slow (roughly three times slower at 408). The results are
compiled in Table 2 and Fig. 2.
First, reactions were carried out starting from a benzylidene derivative of barbituric
acid and a given aldehyde. In all cases, aldehyde exchange was observed even in the
absence of a secondary amine base. However, addition of 10 mol-% molar of a base
increased significantly the exchange kinetics. For instance, the half-life for exchange,
t1/2, was reached up to 85 times faster upon addition of l-proline, compared to the
control experiment without added base (reaction between 2a and 1b; Table 2 and
Fig. 2). For the same reaction, piperidine reduced t1/2 by a factor of 2 only. There was
almost no effect with 2c from 4-nitrobenzaldehyde (1c), because the hydrolysis was
already very fast, in the absence of base. The benzylidene compounds formed from
benzaldehyde (1b) and 4-(dimethylamino)benzaldehyde (1a) were much more
sensitive to base catalysis.
Next, similar reactions were carried out starting from benzylidene derivatives of
malononitrile (M; Table 2, Entry 4). l-Proline was again the most efficient catalyst. For
the exchange reaction between 3b and 1c, it accelerated the reaction 2.5 times.
Piperidine also accelerated the reactions, notably for the reaction between 3c and 1b
(threefold compared to the blank). No exchange was observed over 55 h when starting
from compound 3a, which can be explained by its very low reactivity towards hydrolysis
even in presence of catalytic amounts of l-proline.
The mechanism of the catalysis of the Knoevenagel exchange reaction by a
secondary amine may be considered to involve addition on the benzylidene conjugate
leading to an iminium intermediate by an additionꢀelimination process, followed by
condensation of the released barbiturate or malonate anion with the added aldehyde
(Scheme 3). This catalytic cycle of proline-catalyzed reactions has been extensively
Scheme 2. Reversible Knoevenagel Reactions Involving Exchange of the Aldehyde Moiety between a
Benzylidene Derivative of Barbituric Acid, 2a – 2c (top), and a Malononitrile, 3a – 3c (bottom), and a
para-Substituted Benzaldehyde, 1a – 1c

1224
Helvetica Chimica Acta – Vol. 97 (2014)
Table 2. Kinetic and Equilibrium Thermodynamic Parameters for the Knoevenagel Exchange Reaction
between Benzylidene Compounds 2a – 2c or 3a – 3c and the Aldehydes 1a – 1c. Half-life t1/2. Proportions
[%] of the different compounds at equilibrium in (D6)DMSO/D2O 99 : 1 at 60o
DGob
)
DGob
)
a
b
c
b
b
Entry Reaction
)
Compound distribution [%]
t1/2 [h] Kb
)
Kc
)
1
2a
1b
2b
1a
B
2a þ 1b
fb
49
15
15
16
15
16
16
17
14
8
9
9
9
12
12
12
13
85
1
80
8
0.18
0.18 4.25
0.12 5.47
0.86 0.55
0.53 1.44
4.25
fc 49
rb 49
rc 49
2b þ 1a
2
2a
1c
2c
1a
B
2a þ 1c
fb
31
27
25
27
26
9
9
9
9
10
11
11
11
23
22
23
23
36
20
39
10
0.11
5.25
0.37
1.57
fc 32
rb 31
rc 32
2c þ 1a
3
2b
1c
2c
1b
B
2b þ 1c
fb
29
17
16
15
15
27
29
30
30
14
13
13
14
13
12
11
11
51
4
22
4
0.80
0.56
fc 29
rb 31
rc 31
2c þ 1b
4
3b
1c
3c
1b
M
3b þ 1c
fb
29
22
21
20
22
21
21
21
22
15
16
17
15
13
13
14
12
5
2
15
4
fc 29
rb 28
rc 29
3c þ 1b
a) fb, Forward reaction without catalyst; fc, forward reaction with catalyst 10 mol-% l-proline; rb, reverse
(backward) reaction without catalyst; rc, backward reaction with 10 mol% l-proline as catalyst. b) Kb,
Equilibrium constant of the blank reaction. c) Kc, Equilibrium constant of the catalyzed reaction.
e
d) DGb8 ¼ ꢀRT lnKb [kJmolꢀ1]. ) DGc8 ¼ ꢀRT lnKc [kJmolꢀ1].
documented in [18] (sequential hydrolysis, then condensation). Generally, l-proline
methyl ester proved to be a less efficient catalyst than free proline, again evidencing
participation of the carboxylic acid in the catalytic process. Although the secondary
amine is the key moiety that is involved in the mechanism of the retro-Knoevenagel
reaction, the iminium formed with the benzylidene derivative is likely to be stabil-
ized by the neighboring carboxylate group, which cannot occur with the methyl ester
analog.
2.3. Organocatalysis of Benzylidene/Benzylidene C¼C/C¼C Cross-Exchange Proc-
esses. Recently, we reported on the generation of DCLs involving imine/imine C¼N/
C¼N and benzylidene/imine C¼C/C¼N exchange processes, and on exploiting l-
proline as an organocatalyst [17]. To further extend the range of DCLs, we investigated
the benzylidene/benzylidene C¼C/C¼C cross-exchange reaction between benzylidene-
barbiturates, 2a – 2e, and differently substituted benzylidene-malononitriles, 3a – 3e,
as depicted in Scheme 4. According to the previous results of the benzylidene/imine
C¼C/C¼N exchanges [17], the benzylidene-barbiturates were easily hydrolyzed in

Helvetica Chimica Acta – Vol. 97 (2014)
1225
!
~
Fig. 2. Secondary-amine catalyzed exchange reactions between 2a and aldehyde 1b. , No base;
,
&
*
piperidine; , l-proline methyl ester; , l-proline. The reactions were carried out in triplicate and the
standard deviation was < 5%. The curves were drawn through the experimental points for clarity.
Scheme 3. Mechanism of the Secondary Amine-Catalyzed Knoevenagel Aldehyde Exchange Reaction

1226
Helvetica Chimica Acta – Vol. 97 (2014)
Scheme 4. The C¼C/C¼C Cross-Exchange Reactions between Benzylidene-barbiturates 2a – 2e and
Differently Substituted Benzylidene-malononitriles 3a – 3e in (D6)DMSO at 608
(D6)DMSO/D2O 99 :1 solution, so that pure (D6)DMSO (given as containing less than
0.02% of H2O) was chosen as solvent. Therefore, the further exploration of
benzylidene/benzylidene, C¼C/C¼C, exchanges were conducted in this solvent in
order to minimize hydrolysis, which, however, cannot be fully avoided.
The reactions were carried out with stoichiometric amounts of benzylidene
derivative (12.8 mm each) in pure (D6)DMSO, as well as in the presence of 10% l-
proline. The corresponding t1/2 values and compound distributions of the final solutions
are compiled in Table 3. Equilibrium parameters were not calculated, due to the
occurrence of of hydrolysis to variable extents.
Here again, the efficiency of benzylidene exchange depended on the head-group
substituent of both starting compounds acting as Michael acceptors. Considering first
the pair of compounds 2a, bearing electron-donating group (EDG), and 3c, bearing an
electron-withdrawing group (EWG), the cross-exchange reaction was expected to
yield the products 2c and 3a (Table 3, Entry 1). The 1H-NMR spectrum of the mixture
after the start of the interchange reaction showed new signals at 4.73, 4.94, 5.32, 6.21,
7.26, 7.63, 7.76, 7.91, 9.50, and 12.2 ppm, which were assigned to be the Michael-type
adducts XI and XII (cf. Scheme 5). In addition, the electrospray high-resolution mass
spectrum (HR-ESI-MS) also supported the formation of adducts XI and XII, which
correspond to the peaks at m/z 338.343 ([(M þ H þ Na) þ CH3OH þ H2O]þ) and
360.324 ([(M þ H) þ CH3OH]þ), respectively. Neither adduct could be isolated. In the
reaction mixture, the most reactive electrophiles [19] are 2c and 3c, so that they may
react with the better nucleophile, the manolonitrile anion formed from 3c (pKa(VI) >
pKa(V)), affording adducts XI and XII. Thus, product 2c was formed, but it was
trapped as its Michael-type adduct XII. On the other hand, reaction of malononitrile
with 3c precludes the formation of the product 3a. The possible mechanism of this
cross-exchange reaction is outlined in Scheme 5.
The reversibility of this reaction was studied by combining 2c and 3a (Table 3,
Entry 2). Both starting compounds were fully hydrolyzed, and the anticipated products
2a and 3c did not form even after 20 d, at which time decomposition of the compounds
had occurred. This result indicated that hydrolysis precludes the recondensation to

Helvetica Chimica Acta – Vol. 97 (2014)
1227
Scheme 5. Possible Mechanism of the Cross-Exchange Reaction between Knoevenagel Compounds,
Following Sequential Hydrolysis/Condensation Steps [18]
form 2a and 3c and thus the exchange reaction. For the same reason, no Michael-type
adduct XII was observed.
Next, 2b and 3c were selected as starting compounds, both bearing EWGs (Table 3,
Entry 3). These substrates were more suitable for the cross-exchange than the
molecules exhibiting a large difference in their electrophilicity, i.e., 2a and 3c, and both
products 2c and 3b were formed. The hydrolysis products were observed in a less than
10% yield. In addition, the signals of the Michael-type adducts XI and XII were also
detected in the 1H-NMR spectrum of the mixture as characteristic doublets at 4.94 and
5.32 ppm, attributed to XI (1%), and at 4.72 and 6.20 ppm, attributed to XII (9%), in
agreement with reported values [20] and with MS data (see below). The reverse
reaction was performed by mixing compound 2c and 3b (Table 3, Entry 4), and
component 2b and 3c were obtained in comparable amounts as in the forward case,

1228
Helvetica Chimica Acta – Vol. 97 (2014)
Table 3. Kinetic and Equilibrium Thermodynamic Parameters for Knoevenagel Cross-Exchange Reactions (Scheme 3).
Half-life, t1/2; ꢀ¼ reaction too slow to determine the parameter. Proportions [%] of the different compounds in the C¼C/
C¼C cross-exchange reaction between benzylidene-barbiturates and benzylidene-malononitriles in (D6)DMSO at 608 a,
4-(dimethylamino)benzaldehyde; b, benzaldehyde; c, 4-nitrobenzaldehyde; d, 4-methoxybenzaldehyde; e, thiophene-2-
carbaldehyde; f, 4-(dimethylamino)benzaldehyde hydrate; g, benzaldehyde hydrate; h, 4-nitrobenzaldehyde hydrate; i, 4-
methoxybenzaldehyde hydrate; j, thiophene-2-carbaldehyde hydrate; B, barbiturate; M, malononitrile, and XI and XII,
Michael-type adducts. All measurements were repeated three times. The reproducibility of the values obtained was ꢁ 2%.
Entrya)
t1/2 [h]b)
Reaction
time [h]c)
Compound distribution [%]d)
Starting compound 2a þ 3c
2a
3c
2c
3a
11
13
a
c
f
h
B
M
XI
XII
1b
1c
50
45
164
144
35
33
30
26
5
5
1
2
6
4
–
–
–
–
3
4
3
3
–
2
6
7
Starting compound 2c þ 3a
2c
3a
2a
3c
a
c
f
h
B
M
XI
XII
2b
2c
n.a.
n.a.
142
142
33
32
50
52
–
–
–
–
< 1
< 1
12
9
< 1
< 1
1
3
2
2
1
1
–
–
–
–
Starting compound 2b þ 3c
2b
3c
2c
3b
b
c
g
h
B
M
XI
XII
3b
3c
8
3
53
24
16
12
10
9
22
20
31
26
1
8
7
3
< 1
< 1
1
2
3
5
3
5
< 1
1
5
9
Starting compound 2c þ 3b
2c
3b
2b
3c
b
c
g
h
B
M
XI
XII
4b
4c
20
14
55
32
22
23
33
36
14
13
7
7
1
1
9
6
< 1
< 1
2
2
3
3
3
3
–
–
5
5
Starting compound 2a þ 3d
2a
3d
2d
3a
a
d
f
i
B
M
XI
XII
5b
5c
n.a.
n.a.
9
23e)
488f)
29
49
35
24
49
37
24
1
12
19
1
12
24
–
2
1
–
1
3
–
–
< 1
–
–
1
–
< 1
2
–
< 1
< 1
–
–
1
–
–
–
Starting compound 2d þ 3a
2d
3a
2a
3d
a
d
f
i
B
M
XI
XII
6b
6c
n.a.
n.a.
30
95e)
533f)
98
48
43
23
49
48
30
1
1
19
< 1
2
19
1
1
< 1
< 1
5
4
–
–
< 1
–
–
1
–
1
3
–
–
< 1
–
–
< 1
–
–
–
Starting compound 2e þ 3d
2e
3d
2d
3e
d
e
i
j
B
M
XI
XII
7b
7c
40
2
128
6
46
43
43
40
6
6
6
7
1
2
–
–
< 1
< 1
–
–
–
1
< 1
1
–
1
–
–
Starting compound 2d þ 3e
2d
3e
2e
3d
d
e
i
j
B
M
XI
XII
8b
8c
180
2
532
6
8
6
8
8
42
42
40
39
2
3
–
–
–
–
–
–
1
1
< 1
1
–
1
–
–
a) b, Blank reactions, and c, catalyzed reaction with 10 mol-% l-proline as catalyst. b) Half-life; n.a., not applicable.
c) Reaction time indicates the time when no further change in composition could be observed. d) Compound not
observed; < 1, only traces of product detected. e) The reaction started. f) The reaction did not reach equilibrium.

Helvetica Chimica Acta – Vol. 97 (2014)
1229
indicating that the reaction was reversible, and that the equilibrium had been more or
less reached under thermodynamic control. In the reverse case, only Michael-type
1
adduct XII was detected by H-NMR which exhibited a OH signal at 12.18 ppm, and
doublets at 4.74 and 6.20 ppm, in 5% each under these conditions both in the presence
and absence of l-proline. The formation of XII was also confirmed by HR-ESI-MS
showing peaks at m/z 346.34 ([M þ H3O]þ).
Next compounds 2a and 3d, both bearing EDGs, were reacted (Table 3, Entry 5) to
1
give 2d (19%) and 3a (24%) as products at equilibrium after ca. 29 h. The H-NMR
doublets at 4.48 and 6.09 ppm, observed during the exchange process in the presence of
l-proline, are characteristic of the Michael-type adduct XI (1%), whose formation, was
confirmed by HR-MS data with peaks at m/z 287.21 ([M þ 2 H3O]þ). The hydrolysis
products in the presence of l-proline amounted to less than 4%. Equilibrium was
reached after 29 h in the presence of l-proline, but, in its absence, the reaction was
extremely slow, barely showing exchange after 23 h (1% each of the products 2d and
3a). Furthermore, after 20 d, 2d and 3a formed (12% each), indicating that the reaction
in the absence of catalyst was extremely slow. Nevertheless, the uncatalyzed reaction
did not afford any Michael-adduct, but gave a trace amount (1%) of hydrolysis
products. The reverse reaction, starting with 2d and 3a (Table 3, Entry 6), furnished 2a
and 3d in similar amounts as the forward reaction, and reached equilibrium after 98 h in
the presence of l-proline. Again, in the absence of l-proline, the exchange reaction was
significantly slower, with exchange products being formed in only a trace amount (ca.
1% each) after 4 d, with little change after 22 d where only 2a (1%) and 3d (2%) were
observed. For both forward and back reactions (in the presence of l-proline),
hydrolysis products were formed in yields of less than 4%.
Finally, the exchange reaction between 2e and 3d (Table 3, Entry 7), and the reverse
reaction between 2d and 3e (Table 3, Entry 8) yielded the corresponding products in
comparable amounts in the presence of l-proline. These reactions were faster than the
other series of l-proline-catalyzed Knoevenagel cross-exchanges, approaching equili-
brium after 6 h. In addition, the Michael-type adduct XI was observed in trace amount
(1%) for both forward and reverse reactions (characteristic doublets at 4.46 and
6.07 ppm in the 1H-NMR spectrum). The HR-ESI-MS was used to confirm the
formation of the addition product by showing corresponding peak at m/z 287.2 ([M þ
2 H3O]þ). In the absence of l-proline, exchange products could only be observed after
23 h for both forward and reverse reaction. The equilibrium was approached after 128
and 532 h for forward and backward directions, respectively. Interestingly, the reaction
in the absence of l-proline did not afford any Michael-type adducts, only the hydrolysis
products formed. The hydrolysis products for both in presence and in absence of l-
proline amounted to less than 2%. Under these conditions, l-proline demonstrated
marked organocatalysis, as it accelerated the reaction from twofold (Table 3, Entry 3)
up to ca. 89-fold (Table 3, Entry 8), when one compares the blank reaction at the same
reaction time with the completed catalyzed reaction.
2.4. Benzylidene/Imine, C¼C/C¼N, Exchange Process Involving 1,3-Dimethylbar-
bituric Acid. The C¼C/C¼N interconversion was reported in our previous work [17],
implementing Knoevenagel compounds formed from barbituric acid to achieve the
cross-exchange in pure (D6)DMSO. These reactions were accelerated by l-proline.
Herewith, we extend these earlier studies and report the C¼C/C¼N cross-exchange

1230
Helvetica Chimica Acta – Vol. 97 (2014)
between Knoevenagel compounds derived from 1,3-dimethylbarbituric acid, 4a – 4e,
and the same imine compounds, 5a – 5e, as previously used (Scheme 6). The solubility
of the Knoevenagel compounds was screened in different solvents, and they were
particularly very soluble in CDCl3. The cross-exchange reactions could, therefore, be
carried out in CDCl3 (filtered prior to use through a column of basic alumina to remove
HCl and avoid acid catalysis) with stoichiometric amounts of benzylidene-1,3-
dimethylbarbituric acid derivatives and imine compounds (20 mm each). Both the
forward and reverse reactions were followed by 1H-NMR spectroscopy. In view of the
fast rates of the exchange reactions, the rate constants could not be obtained, but the
equilibrium constants were determined instead, as indicated in the Exper. Part and
compiled in Table 4.
In one of the initial C¼C/C¼N cross-exchange experiments, 4a and 5e (Table 4,
Entry 1), both bearing EDGs, were reacted in CDCl3. After 68 min, the percentage of
the exchange products 4e and 5a remained constant, indicating that equilibrium had
been reached. When 4e and 5a were mixed under the same conditions, 4a and 5e were
formed to give after 69 min a mixture containing the same relative percentages of all
four compounds, thus confirming that the system was at thermodynamic equilibrium.
The first set of experiments in Entry 1 of Table 4 indicates that the interconversion
of the Knoevenagel compound and an imine was facile and under thermodynamic
control when the substrates both bear EDGs. We further explored the interconversion
efficiency as a function of substrate electron affinity, choosing next compounds 4b,
bearing EDGs, and 5c, bearing EWGs (Table 4, Entry 2). The interconversion products
in CDCl3, 4c and 5b, were formed immediately after mixing in a ratio that did not
change after 2 min. In addition, the reverse reaction, determined by combining 4c and
5b under the same conditions, also furnished the products 4b and 5c. For this exchange
pair, 4c and 5b were the favored products showing that the latter compounds are the
most thermodynamically stable under these conditions. In this reaction, hydrolysis
products were observed in the trace amount.
Next, we selected the substrates 4b and 5e for the forward reaction, and combined
4e and 5b for the reverse one (Table 4, Entry 3). The electron-donating properties of
the substituent on 4b were expected to be weaker than that on 4a (Table 4, Entry 1).
Scheme 6. The C¼C/C¼N Cross-Exchange Reactions Between Benzylidene-1,3-dimethylbarbituric Acids
4a – 4e and Imines 5a – 5e in CDCl3 at Room Temperature

Helvetica Chimica Acta – Vol. 97 (2014)
1231
Table 4. Proportions [%] of the Different Compounds in the C¼C/C¼N Cross-Exchange Reactions
between Benzylidene-1,3-dimethylbarbituric Acids and Imines in CDCl3 at Room Temperature. a, 4-
(dimethylamino)benzaldehyde; b, benzaldehydes; c, 4-nitrobenzaldehyde; d, 4-methoxybenzaldehyde;
and e, thiophene-2-carbaldehyde
Entry Starting
compounds
Reactiona) Reaction
timeb) [min]
Compound distribution [%]c)
Keq
4a
5e
4e
5a
a
e
c
f
1
2
3
4
5
6
4a þ 5e
f
r
68
69
22
18
21
17
28
33
29
32
< 1
< 1
–
–
–
–
2.45
4e þ 5a
4b
5c
4c
5b
b
f
4b þ 5c
4c þ 5b
f
r
2
2
13
13
14
14
37
37
36
36
< 1
< 1
< 1
< 1
–
–
7.32
92.05
30.43
152.79
7.31
4b
5e
4e
5b
b
e
f
4b þ 5e
4e þ 5b
f
r
60
13
6
5
4
4
44
46
45
45
–
< 1
–
–
–
–
4b
5a
4a
5b
a
b
f
4b þ 5a
4a þ 5b
f
r
11
3
9
7
7
8
43
45
42
41
< 1
< 1
–
–
–
–
4d
5a
4a
5d
a
d
f
4d þ 5a
4a þ 5d
f
r
11
3
5
3
4
3
46
47
45
47
< 1
–
< 1
–
–
–
4d
5b
4b
5d
b
d
f
4d þ 5b
4b þ 5d
f
r
3
2
14
12
14
14
37
38
35
36
< 1
–
< 1
–
–
–
a) f, Forward reaction; r, reverse reaction. b) Reaction time indicates the time when no further change in
composition could be observed. c) Compound distribution values measured 4 – 5 times and averaged
(standard deviation < 2%). –, Compound not observed; < 1, only traces of product detected.
The favored compounds at equilibrium were 4e and 5b as a result of the electronic
effects of the substituents. In this case, for both forward and reverse reaction, the
hydrolysis products formed in trace amount as well.
The substrates for the next experiment were 4b and 5a for the forward, and 4a and
5b for the reverse reaction (Table 4, Entry 4). The forward and reverse reactions
reached equilibrium after 11 and 3 min, respectively. As compared to the reaction in
Entry 2, 5a has a higher electron donating ability than 5c; therefore, the forward
reaction in Entry 4 was slightly slower than the one in Entry 2. Clearly, the efficiency of
Knoevenagel/imine exchange depends on the electronic profile of both substrates.
To explore the exchange reaction between substrates of markedly different electron
affinities, a mixture of 4d and 5a, bearing EWGs and EDGs, respectively, was studied.
Following mixing, the reaction again occurred immediately and reached equilibrium
after 11 min, with only small amounts of the starting materials left (ca. 3%).
Conversely, the backward reaction was performed by combining 4a and 5d to give 4d
and 5a as the exchange products of 3% each at equilibrium. In addition, the favored

1232
Helvetica Chimica Acta – Vol. 97 (2014)
products at equilibrium were 4a and 5d for both forward and backward reaction under
these conditions.
Finally, the reaction of substrates 4d and 5b (Table 4, Entry 6) was examined, where
5b has lower electron-donating ability than 5a (Table 4, Entry 5). The formation of the
products 4b and 5d were instantaneously formed. The reverse reaction was performed
by mixing 4b and 5d, also to give immediately 4d and 5b. Both forward and reverse
reactions gave the same mixture of the four constituents, indicating that thermody-
namic equilibrium has been reached immediately after mixing.
The cross-exchange reactions observed here may proceed, in principle, following
two different mechanisms. A dissociation/recondensation process may well take place
due to the presence of (unobservable) traces of H2O, similarly to the reaction scheme of
the proline-catalyzed reactions conducted in DMSO (cf. Schemes 3 and 5). On the
other hand, the very fast exchange observed in CDCl3, where no aldehyde could be
detected, would indicate a mechanism of metathesis type, whereby the benzylidene-1,3-
dimethylbarbituric acid would function as a neutral organic Lewis acid [20], while the
imine would be able to act as neutral organic nucleophile. Such a metathesis-like
mechanism of C¼C/C¼N interchange would involve the nucleophilic addition of the
imine N-atom at the CH end of (ring)C¼CH bond of the benzylidene-1,3-dimethyl-
barbituric acid to form the benzylidene-1,3-dimethyl barbituric acidꢀiminium adduct
XV (Scheme 7), followed by internal cyclization to a four-membered azetidine ring
intermediate XVI, which then could open to give either the starting compounds back,
or the products XVIII and XIX, resulting from the cross-exchange of the components.
In a related fashion, the formation of a four-membered carbocyclic intermediate has
been found to occur in organocatalysis of Michael addition reactions of aldehydes to
nitro olefins [13j][13k].
Scheme 7. Possible Metathesis-Like Mechanism of the Cross-Exchange between a Knoevenagel-Type
Substrate (C¼C) and an Imine (C¼N)

Helvetica Chimica Acta – Vol. 97 (2014)
1233
3. Conclusions. – We have explored dynamic C¼C bond processes leading to C¼C/
C¼C cross-exchange between Knoevenagel condensation products by a mechanism
involving formation, hydrolysis, exchange, and recondensation steps. These reactions
are reversible, under thermodynamic control, and they can occur even in the absence of
a basic catalyst. The secondary amine l-proline is an efficient catalyst to accelerate the
formation and hydrolysis of Knoevenagel compounds, increasing the rate of the
exchange with aldehydes or malononitriles. Moreover, l-proline is able to accelerate
the rate of cross-Knoevenagel C¼C/C¼C exchange, and Michael-type adducts are
formed during the cross-exchange process. The %-composition of products at
equilibrium depended on the electron affinity of the starting substrates. Remarkably,
the C¼C/C¼N exchange between imines and Knoevenagel derivatives of 1,3-
dimethylbabituric acid was found to be fast and reversible in CDCl3 solution at room
temperature in the absence of catalyst. This feature provides access to efficient
diversity generation via C¼C/C¼N exchange. It may also signify a change in mechanism
to a metathesis-like process, via an azetidine intermediate. The results described herein
enable the creation of DCLs of higher chemical diversity, thus allowing for the
generation by DCC of receptors, dynamic polymers, or biomaterials of increased
complexity.
S. K. thanks the Khon Kaen University (Thailand) and the Franco-Thai Scholarship Program for a
doctoral fellowship, and K. M. thanks the University of Strasbourg for a teaching assistantship. We also
thank the ERC (Advanced Research Grant SUPRADAPT 290585) for financial support.
Experimental Part
General. Reagents. Thiophene-2-carbaldehyde (98%), benzaldehyde (98%), 4-(dimethylamino)-
benzaldehyde (99%), PhCH2NH2 (99%), 4-nitrobenzaldehyde (98%), barbituric acid (99%), 1,3-
dimethylbarbituric acid (99%), and malononitrile (99%) were purchased from Sigma-Aldrich. 4-
Methoxybenzaldehyde (98%) was purchased from Alfa Aesar. 4-Chlorobenzaldehyde (98%) and l-
proline (99%) were purchased from Avocado and Lancaster, and used as received. Deuterated solvents
were purchased from Euriso-Top and used without further purification except CDCl3, which was passed
through a column of activated alumina (aluminium oxide 90 active basic). NMR Spectra: Bruker Avance
400 spectrometer; referenced to the solvent; in ppm, J in Hz. High-resolution (HR) MS: Bruker Micro
TOF mass spectrometer; in m/z (rel. %).
General Procedure for the Synthesis of Imines (cf. [17]). Equimolar amounts of amine and aldehyde
were dissolved in dry CH2Cl2, and MgSO4 was added. The mixture was stirred at r.t. overnight, and the
solid was filtered off. After drying under vacuum, the products were isolated as colored oils or needles.
General Procedure for the Synthesis of the Knoevenagel Condensation Products (cf. [17]). To 30 ml
of anh. EtOH were added barbituric acid or 1,3-dimethylbarbituric acid (3.0 mmol), the corresponding
substituted benzaldehydes (3.0 mmol), and l-proline in a cat. amount (10%). The soln. was refluxed for
3 – 4 h. The Knoevenagel condensation products that precipitated out of the cooled soln. were filtered and
washed with Et2O, and then dried under vacuum for at least 2 h.
General Procedure for Knoevenagel Formation and Aldehyde Exchange Reactions. Stock solns.
(60 mm; 0.5 ml) of corresponding aldehyde, and barbituric acid or malononitrile (for the formation
studies) or Knoevenagel compounds (for the exchange reaction) were freshly prepared in (D6)DMSO/
D2O 99 :1. For the uncatalyzed reaction, 128 ml of each soln. were added to a NMR tube, followed by
344 ml of (D6)DMSO/D2O 99 :1 to adjust to a final volume of 600 ml. For the catalyzed reaction, 38 ml of a
l-proline stock soln. (12.8 mm in (D6)DMSO/D2O 99 :1) was added to a NMR tube, followed by the
addition of 128 ml of each soln. and 306 ml of (D6)DMSO/D2O 99 :1.

1234
Helvetica Chimica Acta – Vol. 97 (2014)
Thermodynamic and Kinetics Features of Knoevenagel Compound Formation and Aldehyde
Exchange. The half-lives (t1/2) of the reactions were determined by the integration of the Knoevenagel
compound C¼CH and the aldehyde CHO 1H-NMR signals as a function of time and analyzing the curved
obtained. The equilibrium constants are obtained from the amounts of reactants and products according
to Eqn. 1:
K ¼ [2]/[1][B] or K ¼ [3]/[1][M]
(1)
The Gibbs free energy was calculated according to Eqn. 2:
DG ¼ ꢀRT lnKeq [kJ molꢀ1
]
(2)
General Procedure for Knoevenagel C¼C/C¼C Cross-Exchange Reaction. All reactions with two
Knoevenagel compounds were carried out at 608 with a final concentration 12.8 mm and a total volume of
600 ml. Stock solns. of Knoevenagel compounds in (D6)DMSO were prepared. The stock solns. of
reactants were 60 mm and had a total volume of 0.5 ml. The stock soln. of l-proline was 12.8 mm and had
a total volume of 0.5 ml. For the uncatalyzed reaction, 128 ml of the soln. of each Knoevenagel compound
was added to a NMR tube, followed by 344 ml of (D6)DMSO to adjust to a final volume of 600 ml. For the
catalyzed reaction, 38 ml of a l-proline stock soln. (12.8 mm in (D6)DMSO) was added to a NMR tube,
followed by the addition of 128 ml of soln. of each Knoevenagel compound and 306 ml of (D6)DMSO.
General Procedure for C¼C/C¼N Exchange. Stock solns. (40 mm, 0.4 ml) of Knoevenagel compounds
were prepared in CDCl3. For each reaction, 350 ml of each soln. was added to a NMR tube to obtain a
final volume of 700 ml. The final soln. was 20 mm in each component. The %-composition of the reactions
at equilibrium was determined by integration of the Knoevenagel compound C¼CH and the imine N ¼
1
CH H-NMR signals, and the corresponding equilibrium constants K were calculated.
5-[4-(Dimethylamino)benzylidene]pyrimidine-2,4,6(1H,3H,5H)-trione (2a) [17]. Yield: 0.620 g
(80%). Orange solid. M.p. 259 – 2608 ([21]: 262 – 2638]). 1H-NMR (400 MHz, (D6)DMSO): 3.12 (s,
6 H); 6.80 (d, J ¼ 9.3, 2 H); 8.15 (s, 1 H); 8.43 (d, J ¼ 9.3, 2 H); 10.92 (s, 1 H); 11.05 (s, 1 H).
5-Benzylidenepyrimidine-2,4,6(1H,3H,5H)-trione (2b) [21]. Yield: 0.261 g (40%). White solid.
1H-NMR (400 MHz, (D6)DMSO): 7.45 – 7.56 (m, 3 H); 8.08 (d, J ¼ 7.6, 2 H); 8.28 (s, 1 H), 11.23 (s, 1 H);
11.39 (s, 1 H).
5-(4-Nitrobenzylidene)pyrimidine-2,4,6(1H,3H,5H)-trione (2c) [21]. Yield: 0.72 g (92%). Pale-
yellow solid. 1H-NMR (400 MHz, (D6)DMSO): 8.02 (d, J ¼ 8.8, 2 H); 8.25 (d, J ¼ 9.0, 2 H); 8.33 (s, 1 H);
11.32 (s, 1 H); 11.49 (s, 1 H).
5-(4-Methoxybenzylidene)pyrimidine-2,4,6(1H,3H,5H)-trione (2d) [17]. Yield: 0.622 g (90%).
1
Yellow solid. H-NMR (400 MHz, (D6)DMSO): 3.88 (s, 3 H); 7.07 (d, J ¼ 9.1, 2 H); 8.25 (s, 1 H); 8.37
(d, J ¼ 8.9, 2 H); 11.17 (s, 1 H); 11.29 (s, 1 H).
5-[4-(Thiophen-2-yl)benzylidene]pyrimidine-2,4,6(1H,3H,5H)-trione (2e) [17]. Yield: 0.580 g
(87%). Brown-yellow solid. M.p. 2758 ([21]: 270 – 2718]). 1H-NMR (400 MHz, (D6)DMSO): 7.36 (d,
J ¼ 5.0, 1 H); 8.18 (d, J ¼ 3.9, 1 H); 8.28 (d, J ¼ 4.6, 1 H); 8.57 (s, 1 H); 11.26 (s, 1 H); 11.30 (s, 1 H).
2-[4-(Dimethylamino)benzylidene]propanedinitrile (3a) [22]. Yield: 0.410 g (70%). Orange solid.
M.p. 177 – 1818 ([22]: 180 – 1828]). 1H-NMR (400 MHz, (D6)DMSO): 3.10 (s, 6 H); 6.86 (d, J ¼ 9.3, 2 H);
7.84 (d, J ¼ 9.4, 1 H); 8.05 (s, 1 H).
2-Benzylidenepropanedinitrile (3b) [23][24]. Yield: 0.128 g (83%). White solids. M.p. 84 – 858 ([23]:
1
83.5 – 848]). H-NMR (400 MHz, (D6)DMSO): 7.61 – 7.72 (m, 3 H); 7.96 (d, J ¼ 7.8, 2 H); 8.56 (s, 1 H).
2-(4-Nitrobenzylidene)propanedinitrile (3c) [24][25]. Yield: 0.49 g (82%). Yellow solid. M.p. 157 –
1
1588 ([25]: 159 – 1608]). H-NMR (400 MHz, (D6)DMSO): 8.13 (d, J ¼ 9.1, 2 H); 8.44 (d, J ¼ 8.6, 2 H);
8.73 (s, 1 H).
2-(4-Methoxybenzylidene)propanedinitrile (3d) [26]. Yield: 0.45 g (80%). Pale-yellow solid.
1H-NMR (400 MHz, (D6)DMSO): 3.89 (s, 3 H); 7.20 (d, J ¼ 8.8, 2 H); 7.98 (d, J ¼ 9.1, 2 H); 8.41 (s, 1 H).
2-[4-(Thiophen-2-yl)benzylidene]propanedinitrile (3e) [27]. Yield: 0.191 g (40%). Brown-yellow
solid. 1H-NMR (400 MHz, (D6)DMSO): 7.40 (d, J ¼ 4.1, 1 H); 7.95 (d, J ¼ 3.6, 1 H); 8.30 (d, J ¼ 4.3 Hz,
1 H); 8.74 (s, 1 H).

Helvetica Chimica Acta – Vol. 97 (2014)
1235
5-[4-(Dimethylamino)benzylidene]-1,3-dimethylpyrimidine-2,4,6(1H,3H,5H)-trione (4a) [28][29].
Yield: 0.78 g (90%). Orange solid. 1H-NMR (400 MHz, CDCl3): 3.14 (s, 6 H); 3.38 (s, 3 H); 3.39 (s,
3 H); 6.69 (d, J ¼ 9.5, 2 H); 8.39 (d, J ¼ 9.3, 2 H); 8.42 (s, 1 H).
5-(4-Methoxybenzylidene)-1,3-dimethylpyrimidine-2,4,6(1H,3H,5H)-trione (4b) [28]. Yield: 0.73 g
(88%). Yellow solid. 1H-NMR (400 MHz, CDCl3): 3.38 (s, 3 H); 3.39 (s, 3 H), 3.89 (s, 3 H); 6.96 (d, J ¼
9.0, 2 H); 8.30 (d, J ¼ 8.6, 2 H); 8.50 (s, 1 H).
5-Benzylidene-1,3-dimethylpyrimidine-2,4,6(1H,3H,5H)-trione (4c) [28]. Yield: 0.398 g (54%).
Yellow solid. 1H-NMR (400 MHz, CDCl3): 3.36 (s, 3 H); 3.41 (s, 3 H); 7.43 – 7.53 (m, 3 H); 8.04 (d, J ¼
7.7, 2 H); 8.57 (s, 1 H).
5-(4-Chlorobenzylidene)-1,3-dimethylpyrimidine-2,4,6(1H,3H,5H)-trione (4d) [30]. Yield: 0.373 g
(45%). Yellow solid. 1H-NMR (400 MHz, CDCl3): 3.36 (s, 3 H); 3.41 (s, 3 H); 7.42 (d, J ¼ 8.6, 2 H); 8.01
(d, J ¼ 8.9, 2 H); 8.48 (s, 1 H).
1,3-Dimethyl-5-[4-(thiophen-2-yl)benzylidene]pyrimidine-2,4,6(1H,3H,5H)-trione (4e) [31]. Yield:
0.65 g (87%). Brown-yellow solid. 1H-NMR (400 MHz, CDCl3): 3.40 (s, 3 H); 3.41 (s, 3 H); 7.28 (dd, J ¼
4.0, 1 H); 7.89 (d, J ¼ 5.1, 1 H); 7.96 (d, J ¼ 5.6, 1 H).
4-[(E)-(Benzylimino)methyl]-N,N-dimethylaniline (5a) [17]. Yield: 0.698 g (98%). Pale-yellow
solid. 1H-NMR (400 MHz, CDCl3): 2.99 (s, Me2N, 6 H), 4.74 (s, 2 H); 6.68 (d, J ¼ 8.8, 2 H); 7.21 (m, 1 H);
7.31 (d, J ¼ 4.4, 4 H); 7.64 (d, J ¼ 8.7, 2 H); 8.25 (s, CH¼N).
(E)-N-Benzyl-1-(4-methoxyphenyl)methanimine (5b) [17]. Yield: 2.308 g (93%). Pale-yellow solid.
1H-NMR (400 MHz, CDCl3): 3.88 (s, 3 H); 4.83 (s, 2 H); 6.96 (d, J ¼ 9.1, 2 H); 7.31 – 7.26 (m, 1 H); 7.39 –
7.34 (m, 4 H); 7.76 (d, J ¼ 8.4, 2 H); 8.36 (s, CH¼N).
(E)-N-Benzyl-1-phenylmethanimine (5c) [17]. Yield: 0.463 g (79%). Yellow oil. 1H-NMR
(400 MHz, (D6)DMSO): 4.78 (s, 2 H); 7.25 – 7.30 (m, 1 H); 7.33 – 7.38 (m, 4 H); 7.45 – 7.49 (m, 3 H);
7.78 – 7.81 (m, 2 H); 8.52 (s, CH¼N).
(E)-N-Benzyl-1-(4-chlorophenyl)methanimine (5d) [17]. Yield: 0.592 mg (86%). Yellow oil.
1H-NMR (400 MHz, CDCl3): 4.80 (s, 2 H), 7.38 – 7.25 (m, 7 H), 7.70 (d, J ¼ 8.4, 2 H), 8.34 (s, CH¼N).
(E)-N-Benzyl-1-[4-(2-thienyl)phenyl]methanimine (5e) [17]. Yield: 1.93 g (90%). Yellow oil.
1H-NMR (400 MHz, CDCl3): 4.80 (s, 2 H), 7.15 – 6.96 (m, 1 H), 7.39 – 7.23 (m, 6 H), 7.41 (d, J ¼ 5.0,
1 H), 8.46 (s, CH¼N).
REFERENCES
[1] J.-M. Lehn, Chem. Soc. Rev. 2007, 36, 151.
[2] J.-M. Lehn, Chem. – Eur. J. 1999, 5, 2455; A. V. Eliseev, J.-M. Lehn, Science 2001, 291, 2331; O.
Ramstrçm, J.-M. Lehn, Nat. Rev. Drug Disco. 2002, 1, 26.
[3] G. R. L. Cousins, S.-A. Poulsen, J. K. M. Sanders, Curr. Opin. Chem. Biol. 2000, 4, 270; S. Otto,
R. L. E. Furlan, J. K. M. Sanders, Curr. Opin. Chem. Biol. 2002, 6, 321; B. Klekota, B. L. Miller,
Trends Biotechnol. 1999, 17, 205; Nachr. Chem. 2002, 50, 1364.
[4] J.-M. Lehn, in ꢁSupramolecular Science: Where It Is and Where It Is Goingꢀ, Eds. R. Ungaro, E.
Dalcanale, Kluwer, Dordrecht, 1999, p. 287; J.-M. Lehn, Polym. Int. 2002, 51, 825; Polyacylhy-
drazones: W. G. Skene, J.-M. Lehn, Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 8270.
[5] I. Huc, J.-M. Lehn, Proc. Natl. Acad. Sci. U.S.A. 1997, 94, 2106; J. R. Nitschke, J.-M. Lehn, Proc. Natl.
Acad. Sci. U.S.A. 2003, 100, 11970; R. Nguyen, I. Huc, Chem. Commun. 2003, 8, 942.
[6] P. Wipf, S. G. Mahler, K. Okumura, Org. Lett. 2005, 7, 4483.
[7] P. A. Brady, R. P. Bonar-Law, S. J. Rowan, C. J. Suckling, J. K. M. Sanders, Chem. Commun. 1996,
319.
[8] O. Ramstrçm, J.-M. Lehn, ChemBioChem 2000, 1, 41; S. Otto, R. L. E. Furlan, J. K. M. Sanders, J.
Am. Chem. Soc. 2000, 122, 12063; H. Hioki, W. C. Still, J. Org. Chem. 1998, 63, 904.
[9] P. G. Swann, R. A. Casanova, A. Desai, M. M. Frauenhoff, M. Urbancic, U. Slomczynska, A. J.
Hopfinger, G. C. Le Breton, D. L. Venton, Biopolymers 1996, 40, 617.
[10] I. Nakazawa, S. Suda, M. Masuda, M. Asai, T. Shimizu, Chem. Commun. 2000, 881.
[11] C. Brꢄndli, T. R. Ward, Helv. Chim. Acta 1998, 81, 1616; K. C. Nicolaou, R. Hughes, S. Y. Cho, N.
Winssinger, C. Smethurst, H. Labischinski, R. Enderman, Angew. Chem., Int. Ed. 2000, 39, 3823.

1236
Helvetica Chimica Acta – Vol. 97 (2014)
[12] P. J. Boul, P. Reutenauer, J.-M. Lehn, Org. Lett. 2005, 7, 15; P. Reutenauer, E. Buhler, P. J. Boul, S. J.
Candau, J.-M. Lehn, Chem. – Eur. J. 2009, 15, 1893; P. Reutenauer, P. J. Boul, J.-M. Lehn, Eur. J. Org.
Chem. 2009, 1691.
[13] a) A. Erkkilꢄ, I. Majander, P. M. Pihko, Chem. Rev. 2007, 107, 5416; b) B. List, Chem. Rev. 2007, 107,
5413; c) D. A. Bock, C. W. Lehmann, B. List, Proc. Natl. Acad. Sci. U. S.A. 2010, 107, 20636; d) C.
Chandler, P. Galzerano, A. Michrowska, B. List, Angew. Chem., Int. Ed. 2009, 48, 1978; e) J. Seayad,
B. List, Org. Biomol. Chem., 2005, 3, 719; f) S. Mukherjee, J. W. Yang, S. Hoffmann, B. List, Chem.
Rev. 2007, 107, 5471; g) A. B. Northrup, I. K. Mangion, F. Hettche, D. W. C. MacMillan, Angew.
Chem., Int. Ed. 2004, 43, 2152; h) A. B. Northrup, D. W. C. MacMillan, J. Am. Chem. Soc. 2002, 124,
6798; i) B. List, Tetrahedron 2002, 58, 5573; j) J. Bures, A. Armstrong, D. G. Blackmond, J. Am.
Chem. Soc. 2012, 134, 6741; k) D. Seebach, X. Sun, C. Sparr, M.-O. Ebert, W. B. Schweizer, A. K.
Beck, Helv. Chim. Acta 2012, 95, 1064.
[14] B. List, Angew. Chem., Int. Ed. 2010, 49, 1730.
[15] a) E. Knoevenagel, Ber. Dtsch. Chem. Ges. 1896, 29, 172; E. Knoevenagel, Ber. Dtsch. Chem. Ges.
1898, 31, 738; E. Knoevenagel, Ber. Dtsch. Chem. Ges. 1898, 31, 2585; E. Knoevenagel, Ber. Dtsch.
Chem. Ges. 1898, 31, 2596; b) L. F. Tietze, U. Beifuss, in ꢁComprehensive Organic Synthesisꢀ, Ed.
B. M. Trost, Pergamon, Oxford, 1991.
[16] O. Kaumanns, R. Lucius, H. Mayr, Chem. – Eur. J. 2008, 14, 9675.
[17] N. Wilhelms, S. Kulchat, J.-M. Lehn, Helv. Chim. Acta 2012, 95, 2635.
[18] a) P. I. Dalko, L. Moisan, Angew. Chem., Int. Ed. 2004, 43, 5138; b) M. J. Gaunt, C. C. C. Johansson,
A. McNally, N. T. Vo, Drug Discov. Today 2007, 12, 8; c) C. F. Bernasconi, G. D. Leonarduzzi, J. Am.
Chem. Soc. 1980, 102, 1361.
[19] T. Lemek, H. Mayr, J. Org. Chem. 2003, 68, 6880.
[20] F. Seeliger, S. T. A. Berger, G. Y. Remennikov, K. Polborn, H. Mayr, J. Org. Chem. 2007, 72, 9170.
[21] C. S. Reddy, A. Nagaraj, P. Jalapathi, Indian J. Chem., Sect. B 2007, 46, 660.
[22] X.-C. Yang, H. Jiang, W. Ye, Synth. Commun. 2012, 42, 309.
[23] B. B. Corson, R. W. J. Stoughton, J. Am. Chem. Soc. 1928, 50, 2825.
[24] S. Wang, Z. Ren, W. Cao, W. Tong, Synth. Commun. 2001, 31, 673.
[25] H. G. Sturz, C. R. Noller, J. Am. Chem. Soc. 1949, 71, 2949.
[26] Y. Lv, X. Yan, L. Yan, Z. Wang, J. Chen, H. Deng, M. Shao, H. Zhang, W. Cao, Tetrahedron 2013, 69,
4205.
[27] Y. Bouazizi, K. Beydoun, A. Romdhane, H. Ben Jannet, H. Doucet, Tetrahedron Lett. 2012, 53, 6801.
[28] M. C. Rezende, I. Almodovar, Magn. Reson. Chem. 2012, 50, 266.
[29] M. K. Haldar, M. D. Scott, N. Sule, D. K. Srivastava, S. Mallik, Bioorg. Med. Chem. Lett. 2008, 18,
2373.
[30] M. L. Deb, P. J. Bhuyan, Tetrahedron Lett. 2005, 46, 6453.
[31] A. Pałasz, Monatsh. Chem. 2012, 143, 1175.
Received May 28, 2014