Stereoselectivity and kinetics of [4+2] cycloaddition reaction of cyclopentadiene to para-substituted E-2-arylnitroethenes

Article abstract of DOI:10.1002/poc.1853

In spite of diversified electrophilicity of E-2-arylnitroethenes, their [4+2] cycloaddition reactions with cyclopentadiene leads to the corresponding 6-endo-aryl-5-exo-nitronorbornenes and 6-exo-aryl-5-endo-nitronorbornenes as the only reaction products. Stereoselectivity, substituent and solvent effects, and activation parameters, suggest that these reactions occur via a synchronous concerted mechanism on both competing pathways. The experimental results obtained are consistent with the data from B3LYP/6-31G(d) calculations. Due to high electrophilicity of E-2-arylnitroethenes, the reactions studied should be considered as polar [4+2] cycloadditions. Copyright

Full text of DOI:10.1002/poc.1853

Research Article
Received: 4 November 2010,
Revised: 1 February 2011,
Accepted: 5 February 2011,
Published online in Wiley Online Library: 23 March 2011
(wileyonlinelibrary.com) DOI 10.1002/poc.1853
Stereoselectivity and kinetics of [4 þ 2]
cycloaddition reaction of cyclopentadiene
to para-substituted E-2-arylnitroethenesy
a
a
a
*
´
´
Radomir Jasinski , Magdalena Kwiatkowska and Andrzej Baranski
In spite of diversified electrophilicity of E-2-arylnitroethenes, their [4 R 2] cycloaddition reactions with cyclopenta-
diene leads to the corresponding 6-endo-aryl-5-exo-nitronorbornenes and 6-exo-aryl-5-endo-nitronorbornenes as the
only reaction products. Stereoselectivity, substituent and solvent effects, and activation parameters, suggest that
these reactions occur via a synchronous concerted mechanism on both competing pathways. The experimental
results obtained are consistent with the data from B3LYP/6-31G(d) calculations. Due to high electrophilicity
of E-2-arylnitroethenes, the reactions studied should be considered as polar [4 R 2] cycloadditions. Copyright
ß 2011 John Wiley & Sons, Ltd.
Keywords: [4 þ 2] cycloaddition; cyclopentadiene; kinetic; mechanism; nitroalkene; stereoselectivity
The stereospecific reaction course (paths A and B), may
INTRODUCTION
mean that the formation of new s-bonds in the transition states
is concerted, while the non-stereospecific (paths C and D),
suggest a stepwise mechanism. Our aim in this study was to
look for correlations between the packing modes of aryl and
nitro groups in E-2-arylnitroethenes 2a–g and their spatial
arrangement in the [4 þ 2] cycloadducts.
The [4 þ 2] cycloaddition reaction of simple alkenes proceeds via
a concerted mechanism.[2–5] However, in the case of p-deficient
dienophiles and/or when one of the reaction centers is strongly
shielded by a bulky substituent, a zwitterionic mechanism may
compete with the concerted one.[6–8] The zwitterionic reaction
course is usually controlled by the dominant electrophile–
nucleophile interactions.[9,10] Then the cycloaddition reactions
loose their stereospecificity,[6,8,11,12] their activation entropy
assumes low negative values,[11] and their kinetics becomes
highly sensitive to substituent and solvent effects.[13–15]
The competition between the zwitterionic and the concerted
mechanism is an important problem both from a theoretical
and a practical point of view. Therefore, while continuing our
studies[1,16–19] on the reactivity of conjugated nitroalkenes
in [4 þ 2] cycloadditions, in this work we evaluate the
stereospecificity and kinetics of the reaction of cyclopentadiene
(1) with a homogenous series of E-2-arylnitroethenes 2a–g.
In addition to the experiments, we have performed quantum-
chemical simulations of the reaction paths for model reactions.
In dienophiles 2a–g, the degree of shielding of the double bond
is very similar, but their electrophilicity (v) is different (Table 1).
According to Domingo’s terminology,[9,20] they can be considered
as strong electrophiles (v > 1.5 eV), therefore their reactions
with cyclopentadiene should have an enhanced polar or even
zwitterionic character. The approach presented here sheds light on
the reaction mechanism, which to the best of our knowledge, has
not so far been the subject of detailed investigations.
The reactions of 1 with E-2-arylnitroethenes 2a, 2e, and
2g have been reported,[25–28] but without analysis of the
stereochemical outcomes, because separation of the reaction
mixtures into their component stereoisomers was not carried
out. Moreover, the cited works also lack detailed information
about the reaction conditions. We have carried out a detailed
study of the reaction stereospecificity with experiments aimed at
the determination of optimal reaction conditions and methods
of separation of the cycloadducts. We have found that the
cycloaddition reactions are best carried out in sealed ampoules at
60–80 8C, using a 3-fold molar excess of the diene and an aprotic
solvent as the reaction medium. Under such conditions, the
!
E-2-arylnitroethenes 2a–g did not undergo E Z isomerization,
while after 48 h of heating at 80 8C, the conversion of dienophile
always exceeded 90%. Detailed HPLC and 1H-NMR analysis of
the residue left after vacuum removal of the solvent and the
excess of cyclopentadiene confirmed that regardless of the
type of substituent R in aryl ring of the dienophile, two new
compounds were formed with retention times (Rt) different
from the retention times of substrates. For example, the
product mixture from the reaction 1 þ 2a contained two
´
*
Correspondence to: R. Jasinski, Institute of Organic Chemistry and Technology,
Cracow University of Technology, ul. Warszawska 24, 31-155 Cracow, Poland.
E-mail: radomir@chemia.pk.edu.pl
RESULTS AND DISCUSSION
Stereospecificity of the reaction of cyclopentadiene with
E-2-arylnitroethenes
´
´
a
R. Jasinski, M. Kwiatkowska, A. Baranski
Institute of Organic Chemistry and Technology, Cracow University of
Technology, ul. Warszawska 24, 31-155 Cracow, Poland
Part 6 of the series ‘‘Conjugated Nitroalkenes in Cycloaddition Reactions’’; Part
5 see Ref.[1]
y
Depending on the reaction mechanism, we could expect four
[4 þ 2] cycloadducts in the reactions studied (Scheme 1).
J. Phys. Org. Chem. 2011, 24 843–853
Copyright ß 2011 John Wiley & Sons, Ltd.

´
´
R. JASINSKI, M. KWIATKOWSKA AND A. BARANSKI
Table 1. The global electrophilicity and energy of the FMO levels of cyclopentadiene 1 and E-2-arylnitroethenes 2a–g
Compound
1
2a
2b
2c
2d
2e
2f
2g
v (eV)
Dv (eV)
0.83
–
0.63
8.83
3.70
2.87
1.94
9.83
3.03
2.20
–
2.89
2.06
1.40
8.96
2.88
2.05
1.37
9.04
2.65
1.82
1.35
9.12
2.56
1.73
1.19
8.79
2.42
1.59
1.14
8.50
ꢀELUMO (eV)a
ꢀEHOMO (eV)b
–
a Data taken from Refs.[21,22]
b Data taken from Refs.[23,24]
products (P1 i P2) with Rt equal to 29.7 and 33.2 min, respectively
(Fig. 1, II), besides a small quantity of unreacted nitroalkene (S)
and cyclopentadiene (D) dimer.
However, the stereochemistry assignment was possible
with the aid of 1H-NMR spectroscopy. In particular, in 1D
1H-NMR spectra of both isomers, signals from the aromatic
protons (Ha and Hb) appear at the lowest magnetic fields, while
the protons from the methylene bridge (H7 and H8) show up at
the highest fields. The protons on the carbon atoms with sp2
hybridization (H2 and H3) exhibit similar chemical shifts. In
both stereoisomers the signal from the H4 proton appears at a
slightly lower field than that of H1. This results from a stronger
deshielding effect of the nitro group in vicinity of H4 than
the corresponding effect of the aromatic ring in vicinity of H1.
The resonance signal of the proton H6 in the major isomer (P2)
occurs at higher field than the corresponding H6 signal of the
minor isomer (P1). A similar chemical shift difference is observed
in the case of protons H5. However, the H5 signal of the minor
isomer appears at a higher field than that of the major isomer.
This is probably caused by the double bond anisotropy in
the norbornene molecule[36] – the proton in the endo-position
feels the presence of the double bond more than the proton in
the exo-position. Comparison of the coupling constants of the
Isolation of the cycloadducts was possible with the aid of
semipreparative HPLC, but only when the separation process was
carried out at 5 8C (Experimental section). In this way the products
P1 and P2 were obtained in pure enough form to perform their
physicochemical characterization.
In the IR spectra of both products, the vibration bands[29] are
—
characteristic of nitro groups, >C C< bonds, a bridging
—
methylene group and a para-substituted benzene ring (Table 2).
The MS spectra of the products are very similar. The position
of molecular ion (MR) corresponds to the sum of molecular
masses of the substrates 1 i 2a (m/e ¼ 260), which in combination
with the results of elemental analysis (CHN), allow us to assign
the formula C13H12N2O4 to the isolated cycloadducts. The
fragmentation pattern of MR. is
a combination of the
fragmentation patterns characteristic of norbornenes[30–32] and
nitrocycloalkanes,[33–35] but it does not allow differentiation
between the stereoisomers.
Scheme 1. R ¼ NO2 (a), COOCH3 (b), Br (c), Cl (d), H (e), CH3 (f), CH3O (g)
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 843–853

CYCLOADDITION REACTION OF CYCLOPENTADIENE
Figure 1. Chromatogram of the post-reaction mixture of cyclopentadiene 1 with E-2-p-nitrophenyl-nitroethene 2a at 25 8C (I – 70% MeOH was used as
the eluent) and at 5 8C (II– 60% MeOH was used as the eluent)
protons H5 and H6 in 3a–6a (Fig. 2) calculated with PcModel
3.2 algorithm, with those obtained from 1H-NMR spectra, has
confirmed that in both isomers these protons are located on
opposite sides of the norbornene ring (Table 3). Hence the
compounds P1 and P2 can be assigned the structure of 6-endo-
p-nitrophenyl-5-exo-nitronorbornene (3a) and 6-exo-p-nitrophenyl-
5-endo-nitronorbornene (4a), respectively.
Such stereoisomerism of the cycloadducts 3a and 4a is also
consistent with the 2D 1H-NMR spectra. In particular, in the
NOESY spectrum of 3a (Fig. 3), a correlation signal is observed
at the intersection of the spectrum diagonal with the lines
projecting chemical shifts of the protons H6 and H7 onto the
diagonal, indicating proximity of the protons in space. No
correlation signal was observed for the protons H5 and H7.
This confirms that in 3a the proton H6 is in the exo-, while H5 is
in the endo-position. On the other hand, in the case of the
stereoisomer 4a, no through space coupling between H6 and
H7 appears in the NOESY spectrum, while the correlation
between H5 and H7 and between H7 and a-protons of aromatic
ring is clearly visible. This means that the proton H6 is located in
the endo-, while the proton H5 in the exo-position.
cyclopentadiene with parent nitroethene,[38] because both
substituents can interact with the p-system of cyclopentadiene.
Kinetics of the reaction of cyclopentadiene with
E-2-arylnitroethenes
The stereospecificity of the cycloadditions studied suggests that
these reactions occur according to a concerted mechanism on
both competing pathways. However, the stereospecificity alone
cannot be considered as a sufficient proof of the mechanism.
The reactions may proceed via a zwitterionic mechanism, while
preserving the original dienophile configuration in the products,
when the activation barrier for the carbocyclic ring closure in the
postulated intermediates (Scheme 1) is lower than the barrier
for rotation around the C—C bond of the nitroethyl fragment.
Such a scenario is most likely for the reactions 1 þ 2a ! 3a and
1 þ 2a ! 4a, where the difference in electrophilicity of the
substrates (Dv) is the highest (Table 1).
In order to shed more light on the nature of the mechanism,
the kinetics of the reactions were studied. In particular, the effect
of substituents and solvents was studied and the activation
parameters were determined.
It is necessary to emphasize that according to B3LYP/6-31G(d)
calculations, the distances H5–H7 and H6–H7 in the stereoisomer
The kinetic runs were carried out at 60, 70, and 80 8C. They were
followed to 65–80% completion by quantitative HPLC analysis of
the reaction mixture (Experimental section). All reactions were
carried out using 20–22 molar excess of E-2-arylnitroethene
which afforded pseudo-first order rate constants kobs. The overall
second-order rate constants ktotal, (which are the sum of kA and
kB, cf. Scheme 1), were obtained from the quotient of kobs and the
initial concentration of suitable E-2-arylnitroethene. It was found
that the product composition is controlled kinetically, since the
molar ratio g ¼ [3]/[4], was constant throughout the reaction
course. From the overall rate constants (ktotal) and molar ratios (g),
the rate constants kA and kB were evaluated using the formulas
kA ¼ gktotal/(g þ 1) and kB ¼ ktotal/(g þ 1). The results are pre-
sented in Table 4. From these data, the activation parameters
were estimated from Eyring equation[39] in the form:
˚
3a are equal to 3.74 and 2.41 A, while the corresponding
˚
distances in the stereoisomer 4a are equal to 2.57 and 3.75 A,
respectively (Fig. 2).
Hence, the reaction 1 þ 2a proceeds with retention of
the dienophile configuration on both competitive reaction paths
(i.e., the paths A and B in Scheme 1). The reactions of the other
E-2-arylnitroethenes occur in an analogous manner regardless of
the magnitude of p-deficiency of the dienophile (Table 1). In all
product mixtures only two new products were identified, which
on the basis of elemental analysis (CHN), and MS, IR, and 1H-NMR
spectra (Table 2) were assigned the structures of 6-endo-aryl-5-
exo-nitro-norbornenes (3a–g) and 6-exo-aryl-5-endo-nitro-nor-
bornenes (4a–g). The results are in clear agreement with the
predictions based on the Alder rule I (cis-principle of addition) and
II (maximum accumulation of unsaturation).[4,5,37] It was noted at
this point, that presence of two substituents on the ethene
decreases endo-selectivity in comparison to DA reaction of
¼
¼
logðk=TÞ ¼ 10:319 þ DS =4:576ꢀDH =4:576T
(1)
J. Phys. Org. Chem. 2011, 24 843–853
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

´
´
R. JASINSKI, M. KWIATKOWSKA AND A. BARANSKI
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 843–853

CYCLOADDITION REACTION OF CYCLOPENTADIENE
Figure 2. Views of structures of 6-p-nitrophenyl-5-nitronorbornenes 3a–6a
¼
The activation enthalpies (DH ) were estimated from the
slopes of the plots of logarithm of the quotient of kA or kB
and the absolute temperature T versus the reciprocal of absolute
case of sp (Fig. 4):
logkA ¼ 1:60 spꢀ4:40ðR ¼ 0:997Þ
logkB ¼ 1:27spꢀ3:77ðR ¼ 0:994Þ
(2)
(3)
¼
temperature (1/T), while the entropies of activation (DS ) were
determined from the intercepts of these plots.
The effect of the substituent on the reaction rates on paths A
and B was studied first in the usual way by plotting log kA
and log kB against the Hammett substituent constants sþ, s0,
sR, sI, and sp.[39] The best correlations were obtained in the
This means that the effect of the substituent R in the phenyl
ring of E-2-arylnitroethenes 2a–g is transferred to the reaction
center both by resonance and inductive interactions. It is
Table 3. The selected interatomic distances, angles and coupling constants for nitronorbornenes 3–6a
J5,6 Coupling constants
(Hz)
Geometry parametersa
Calc.b
Exp.
˚
rH5H7 (A)
˚
rH6H7 (A)
Isomer
H5—C—C—H6 angle (8)
3a
4a
5a
6a
3.74
2.57
2.53
3.75
2.41
3.75
2.47
3.75
131.2
131.1
6.4
5.7
5.7
10.2
10.1
3.8
4.3
–
2.3
–
a Data taken from B3LYP/6-31G(d) calculations.
b Data taken from PcModel simulations.
J. Phys. Org. Chem. 2011, 24 843–853
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

´
´
R. JASINSKI, M. KWIATKOWSKA AND A. BARANSKI
Figure 3. The NOESY correlations for nitronorbornenes 3a and 4a
interesting to note at this point, that similar linear relationships
exist between log kA and log kB and the values of index v
(Table 1), which were calculated for E-2-arylnitroethenes 2a–g
using the formulas recommended by Parr and Yang.[40] The
correlations are, however, worse:
increase of the rate constants kA and kB, the transition states on
both competitive reaction paths have only slightly polar character,
as indicated by the r-values (r¼ 1.60 and r¼ 1.27), which by no
means proves a zwitterionic reaction mechanism. Moreover,
positive r-values indicate that charge transfer with the transition
state occurs from cyclopentadiene substructure towards the
corresponding nitroalkene substructure. Similar r-value was noted
for the [4 þ 2] cycloaddition reaction of 2,3,4,5-tetraphenyl-
cyclopentadien-1-one with 2-R-phenyl-1-carbomethoxyacetylenes,
whose concerted mechanism is more certain.[41]
logkA ¼ 1:36 vꢀ8:02ðR ¼ 0:963Þ
logkB ¼ 1:08vꢀ6:67ðR ¼ 0:965Þ
(4)
(5)
The linear relationships between log kA or log kB, and the
parameters sp or v suggest a similar structure for the transition
states involved in the reactions studied. Even though replacement
The results obtained are in agreement with the perturbation
interaction diagram (Fig. 5) determined by us on the basis of
experimental values of the FMO energies estimated for the
reactants 1 and 2a,c–g (Table 1).
of the methoxy group (sp CH O ¼ ꢀ0:27) in E-2-arylnitroethene
3
2g with a nitro group (sp NO ¼ 0:78) causes up to a ca. 25-fold
2
Table 4. Results of kinetic measurements of [4 þ 2] cycloaddition of cyclopentadiene 1 with E-2-arylnitroethenes 2a–g
ktotal ꢁ 104
Isomer
t (8C) (dm3/mol s) ratio g (dm3/mol s) (dm3/mol s)
kA ꢁ 104
kB ꢁ 104
Reaction Solvent (ET(30))
R
C
SD
1
2
3
4
5
6
7
8
1 þ 2a
1 þ 2b
1 þ 2c
1 þ 2d
1 þ 2e
Nitromethane (46.3)
70
70
70
70
70
60
70
80
70
70
70
70
70
24.95
9.31
4.86
3.55
2.66
0.88
2.25
4.20
1.89
1.40
1.04
1.14
0.99
0.40
0.34
0.30
0.30
0.18
0.20
7.13
2.36
1.12
0.82
0.41
0.15
0.38
0.70
0.41
0.34
0.25
0.20
0.16
17.82
6.95
3.74
2.73
2.25
0.73
1.87
3.50
1.48
1.06
0.79
0.94
0.83
0.997 0.094 0.066
0.998 0.074 0.046
0.997 0.100 0.047
0.998 0.073 0.023
0.994 0.139 0.032
0.995 0.116 0.035
0.999 0.020 0.020
0.999 0.047 0.014
0.999 0.058 0.025
0.998 0.068 0.038
0.999 0.042 0.018
0.996 0.114 0.048
0.999 0.064 0.072
Ethanol (51.9)
Nitromethane (46.3)
9
10
11
1,2-Dichloroethane (41.9)
Benzene (34.5)
Cyclohexane (31.2)
Nitromethane (46.3)
0.28
0.32
0.31
0.21
0.19
12 1 þ 2f
13 1 þ 2g
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 843–853

CYCLOADDITION REACTION OF CYCLOPENTADIENE
Figure 4. Plot of log k versus Hammet constants sp and electrophilicity index v for the reaction of cyclopentadiene 1 with E-2-arylnitroethenes 2a–g
From the diagram it is obvious that the energy gap between
the HOMO of diene 1 and the LUMO of arylnitroethenes 2a,c–g
predominates in all cases and that this gap (DE1) is smaller than
the gap (DE2) between the HOMO of arylnitroethenes 2a,c–g and
the LUMO of diene 1. Such an orbital arrangement indicates
that arylnitroethenes 2a,c–g participate rather in the normal
[4 þ 2] cycloadditions, according to Sustmann’s classification
of Diels–Alder reactions.[42] Furthermore, it is evident from Fig. 5
that DE1 decreases with increase of sp values. Consequently,
the reactivity should also increase in the same order, which
results in positive r-values in the Hammett Eqns (2) and (3).
In order to explain the solvent effect, we studied kinetics of
1 þ 2e ! 3e and 1 þ 2e ! 4e in solvents with differing ionising
power (Table 4). The Dimroth ET(30) values[14,39] based on
solvatochromism of pyridinium N-phenylbetaines, were used as a
measure of this power. In this investigation, nitroalkene 2e has
been chosen as a model compound, because it is easily soluble in
solvents with ET(30) values varying over a wide range, whereas
nitroalkene 2a is practically insoluble in nonpolar solvents. The
effect of the solvent on the cycloaddition rate constants kA and kB
was examined by plotting log kA and log kB versus the solvent ET
values (Fig. 6). Regression analysis of the results has shown that
there exist linear relationships represented by the following
equations:
logkA ¼ 0:009ETð30Þꢀ4:81ðR ¼ 0:807Þ
logkB ¼ 0:021ETð30Þꢀ4:74ðR ¼ 0:991Þ
(6)
(7)
As it can be seen from the equations, the solvent sensitivity
parameter – represented by the slopes in Eqns (6) and (7) – equal
0.009 for path A and 0.021 for path B. The small response of
the reaction kinetics to variation of the solvent ionizing power
contradicts the formation of zwitterionic intermediates in the
rate-determining step, and at the same time, supports the
concerted mechanism for both competing pathways. Moreover,
the positive sign of the parameter indicates that the transition
states on paths A and B are not much more polar than the
corresponding intermediates, which on the potential energy
surface are usually presented as local minimums.[16,43] This
Figure 5. FMO interaction diagram for [2 þ 3] cycloaddition of cyclopentadiene 1 to E-2-arylnitroethenes 2a,c–g. Data for diagram taken from Table 1
J. Phys. Org. Chem. 2011, 24 843–853
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

´
´
R. JASINSKI, M. KWIATKOWSKA AND A. BARANSKI
Figure 6. Plot of log k versus Dimroth ET(30) constants for the reaction of
cyclopentadiene 1 with E-2-phenylnitroethene 2e
Figure 7. The enthalpy profiles of 1 þ 2a and 1 þ 2g reactions
conclusion is in agreement with the activation parameters
calculated for the reactions 1 þ 2e ! 3e and 1 þ 2e ! 4e. In
Formation of the pre-reaction complex (LM) does not
require passing through any activation barrier, because the
corresponding enthalpy change does not exceed 1.5 kcal/mol. It
has to be pointed out that at 298 K the LMs are exclusively
enthalpic in nature (Table 5), because the entropy factor (TDS)
makes the free energy change (DG) greater than zero, that
excludes existence of LMs as stable intermediates. Moreover,
the distance between the reaction sites in the pre-reaction
¼
particular, the activation enthalpies (DH ) were found to be equal
¼
17.3 and 17.7 kcal/mol, whereas the entropies of activation (DS )
were ꢀ28.6 and ꢀ24.5 cal/mol K, respectively. These values are
typical for concerted [2 þ 4] cycloaddition reactions.[4,5] For
¼
example,[44] DH of the concerted reaction of cyclopentadiene
¼
with ethene equals 22.5 kcal/mol, while DS is ꢀ32.1 cal/mol K. In
˚
complexes is more than 3.5 A, which is far beyond typical bond
the case of the reaction of 1,1-dimethoxy-1,3-butadiene with
tetracyanoethylene,[11] which is known to occur via a zwitterionic
lengths in transition states. None of the LMs has the character of
a CT-complex, because the data calculated from the Leroy’s
equation[46] indicate that no electron transfer between the
substructures occurs (cf. t indices in Table 5).
¼
¼
mechanism, DS is only ꢀ6.2 cal/mol K. The DH data,
determined for the reactions studied, suggest that the
rehybridization of the reaction centers within the transition
states is small. According to Hammond terminology,[45] these are
early transition states. On the other hand, large negative values of
Further transition of the reacting system from the area
of pre-reaction complexes (LM) towards the energy minima
corresponding to the reaction products requires overcoming
the activation barrier. The B3LYP/6-31G(d) calculations indicate
the nitroalkene 2a is the most reactive component of the
cyclopentadiene/nitroalkene system. Regardless of the nitroalk-
ene nature, kinetically most favored direction of the substrates
transformation is always the path B, which leads to the corres-
ponding 6-exo-aryl-5-endo-nitronorbornenes (Table 5). Hence,
the theoretical predictions of the reaction paths perfectly
correlate with the experimental data (Fig. 8).
¼
DS point to a considerable restriction of free movement of the
reactants within the transition states. Hence, the results obtained
support the concertedness of the studied cycloadditions and
contradict the formation of zwitterionic intermediates in their
rate-determining steps.
Quantumchemical analysis of the reaction of
cyclopentadiene with E-2-arylnitroethenes
In addition to the experimental data, we have simulated the
reaction paths for the pairs of substrates with largest and smallest
difference in electrophilicity. We hoped, that such approach
would provide a supportive information about the nature of
critical structures in the title reaction. The quantumchemical
simulations were carried out using B3LYP/6-31G(d) algorithm in
the analogous manner as in the case of previously analyzed DA
reaction of cyclopentadiene with E-2-phenyl-1-cyanonitroethene.[16]
It has turned out that the energy profiles on the paths A and B
are very similar. Between the energy minimum of substrates
(1 þ 2a or 1 þ 2g) and the energy minimum of the cycloadduct,
there is always only one transition state (TS), preceded with
a shallow minimum (LM) corresponding to the pre-reaction
complex (Fig. 7).
The transition states (TS) localized on the PES have biplanar
structure, characteristic for concerted [4 þ 2] cycloaddition
reactions.[3–5] New s-bonds within the TS are formed between
C4—C5 and C1—C6 reaction sites. The former bond C1—C6
always forms faster than the latter one, however, the C4—C5
bond is only slightly less advanced (Table 5). Hence, the degree of
symmetry of the complexes analyzed can be considered as high.
The TS symmetry is determined to some extent by the type
of substituent in the phenyl ring at the reaction site C6. This
is reflected in the set of asymmetry indices (Dl), calculated from
the relationship proposed in the Ref.[16] The indices are 0.06
and 0.09 for the TSA and TSB transition state complexes of the
reaction with more electrophilic nitroalkene (2a), and, 0.03 and
0.04 for the corresponding transition states of the reaction with
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 843–853

CYCLOADDITION REACTION OF CYCLOPENTADIENE
Table 5. Essential molecular properties and Eyring parameters from the B3LYP/6-31G(d) calculations of critical structures for
reaction of cyclopentadiene 1 with 2-arylnitroethenes 2a,g (T ¼ 298 K; DH and DG values are in kcal/mol; DS values are in cal/mol K)
C4—C5
C1—C6
˚
r (A)
˚
r (A)
Reaction
Structure
l
l
Dl
mꢂ(D)
t (e)
DH
DS
DG
1 þ 2a
LMA
TSA
3a
LMB
TSB
4a
LMA
TSA
3g
LMB
TSB
4g
3.827
2.285
1.577
3.762
2.280
1.574
7.306
2.239
1.577
6.680
2.224
1.574
0.551
3.620
2.182
1.587
3.847
2.174
1.597
5.532
2.201
1.585
7.671
2.195
1.594
0.626
0.07
0.52
4.68
5.13
0.63
4.61
4.97
8.96
5.74
4.05
8.36
6.02
4.19
0.00
0.21
0.13
0.01
0.21
0.12
0.03
0.17
0.10
0.03
0.16
0.09
ꢀ0.8
ꢀ27.1
ꢀ49.3
7.3
32.1
17.4
0.552
0.580
0.587
0.639
0.611
0.624
0.09
0.03
0.04
ꢀ0.4
ꢀ24.6
ꢀ49.7
7.0
31.9
17.1
1 R 2g
ꢀ1.4
ꢀ21.3
ꢀ48.2
5.0
35.8
21.5
ꢀ1.3
ꢀ20.8
ꢀ48.9
4.9
35.5
20.9
Figure 8. Views of the transition complexes structures of the reaction of cyclopentadiene 1 with nitroalkenes 2a,g
less electrophilic nitroalkene (2g). Despite high symmetry the
transition state complexes are polar in nature, as indicated by the
magnitude of their dipole moment (m > 4.5D) and charge transfer
index (t ¼ 0.16–0.21e).
the corresponding 6-endo-aryl-5-exo-nitronorbornenes 3a–g
and 6-exo-aryl-5-endo-nitronorbornenes 4a–g as the only
reaction products. The electrophilicity of the dienophiles is
probably too small to favor the zwitterionic path for these
reactions. The results obtained harmonize with the data
from B3LYP/6-31G(d) calculations of competing pathways for
the reaction 1 þ 2a and 1 þ 2g, as well as with the secondary
a-deuterium kinetic isotope effects[18] observed in cycloaddition
of 1 þ 2e. In terms of Domingo’s terminology,[10] the reactions
under study should be considered as polar (P-DA) [4 þ 2]
cycloadditions.
CONCLUSIONS
In spite of diverse p-deficiency of E-2-arylnitroethenes
2a–g, their [4 þ 2] cycloadditions with cyclopentadiene 1 occur
through a synchronous concerted mechanism and lead to
J. Phys. Org. Chem. 2011, 24 843–853
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc

´
´
R. JASINSKI, M. KWIATKOWSKA AND A. BARANSKI
in an ice bath. Samples (250 ml) were taken from the ampoules
with a microsyringe and were diluted with 1000 ml of cold
methanol. The solutions were analyzed by HPLC at the wave-
length 254 nm. The control experiments showed that under
the conditions used for kinetic measurements, the concen-
trations of 3 and 4 were measured with an error smaller than
3–4%. After completion of the reaction, the ratio of the
cycloadducts in the product mixture (g ¼ [3]/[4]) was measured
by HPLC. The analyses were carried out at 5 8C because at this
temperature the peaks corresponding to the cycloadducts were
clearly separated (Fig. 1).
Thirteen series of measurements were completed. Numerical
analysis using the MATCAD 7 program, showed that for all kinetic
runs, satisfying linear relationships existed between ln (A1 ꢀ At )
and time t, according to the pseudo first-order kinetic
equation.[39] The rate constants obtained are collected in Table 4.
EXPERIMENTAL
Melting points were determined on a Boetius apparatus and
are uncorrected. Elemental analyses were determined on a
Perkin–Elmer PE-2400 CHN apparatus. Mass spectra (EI, 70 eV)
were obtained using a Hewlett–Packard 5989B spectrometer. IR
spectra were recorded on a Bio-Rad spectrophotometer. 1H-NMR
spectra were taken on a Bruker AC (300 MHz) spectrometer, using
TMS as an internal standard, and CDCl3 as a solvent. Liquid
chromatography (HPLC) was done using a Knauer apparatus
equipped with a UV–VIS detector (l ¼ 254 nm). For monitoring
of the reaction progress, LiChrospher 100-10-RP column
(4 ꢁ 240 mm) and 60–70% methanol as the eluent at flow
rate 1 ml/min were used. The separation of the post-reaction
mixtures was performed on the same Knauer apparatus, using a
semipreparative column (LiChrospher 100-10-RP, 16 ꢁ 240 mm)
and 60–65% methanol as the eluent at flow rate 10 ml/min.
Acknowledgements
Materials
The authors acknowledge the financial support of these studies
by the Polish Ministry of Science and Higher Education (grant C2/
293/DS/09).
Cyclopentadiene 1 was prepared by pyrolysis of commercially
available dicyclopentadiene (Aldrich) at 180–200 8C, according to
the standard procedure.[47] Just before use it was distilled under
atmospheric pressure, using 25 cm Vigreux column. E-2-Arylnitro-
ethenes 2a–g were obtained by condensation of appropriate
aromatic aldehydes with nitromethane, according to reported
procedures.[48–50] Their purity was confirmed by HPLC analyses.
Pure grade (POCh, Merck) ethanol, nitromethane, 1,2-dichloro-
ethane, benzene, and cyclohexane were used as solvents. Before
usage they were carefully purified according to standard
procedures.[51]
REFERENCES
´
´
´
[1] R. Jasinski, P. Michorczyk, E. Jasinska, O. Kojfman, A. Baranski, Izv.
Vuzov. Khim. Khim. Tekhnol. 2011, 54(5), (in press).
[2] F. Fringuelli, A. Taticchi, Diels–Alder Reaction; Selected Practical
Methods, J. Wiley & Sons, New York, 2002.
[3] F. Fringuelli, A. Taticchi, Dienes in the Diels–Alder Reaction, J. Wiley &
Sons, New York, 1990.
[4] A. Wassermann, Diels–Alder Reactions; Organic Background and Phy-
sico-Chemical Aspects, Elsevier, Amsterdam, 1965.
Cycloaddition of cyclopentadiene (1) with
E-2-arylnitroethenes (2a–g)
[5] A. S. Onishchenko, Diene Synthesis, Israel Program for Scientific
Translations, New York, 1964.
General procedure
´
´
[6] R. Jasinski, M. Kwiatkowska, A. Baranski, Wiad. Chem. 2007, 61,
In a typical procedure, a solution of 1 mmol of an appropriate
E-2-arylnitroethene and 3 mmol of cyclopentadiene in 2 ml of
nitromethane was heated in a sealed ampoule of ca. 3 ml capacity
at 80 8C. The reaction was monitored by HPLC at the wavelength
254 nm. After completion of the reaction, the solvent was
evaporated under vacuum and the residue was separated by
semipreparative HPLC. Under these conditions, for each pair of
substrates, two [4 þ 2] cycloadducts with different Rt were
isolated. Their essential physicochemical properties are pre-
sented in Table 2.
485–514.
[7] S. Lakhdar, F. Terrier, D. Vichard, G. Berionni, N. El-Guesmi,
R. Goumont, T. Boubaker, Chem. Eur. J. 2010, 16, 5681–5690.
[8] J. Sauer, R. Sustmann, Angew. Chem. Int. Ed. 1980, 19, 779–807.
[9] P. Perez, L. R. Domingo, A. Aizman, R. Contreas, in Theoretical Aspects
of Chemical Reactivity, Vol. 19 (Ed.: A. Toro-Labbe), Elsevier Science,
New York, 2007.
[10] L. R. Domingo, J. A. Saez, Org. Biomol. Chem. 2009, 7, 3576–3583.
[11] R. Sustmann, S. Tappanchai, H. Bandmann, J. Am. Chem. Soc. 1996,
118, 12555–12561.
[12] C. S. Kniep, R. Boese, R. Sustmann, Tetrahedron 2000, 56, 4157–4162.
[13] R. Huisgen, Pure Appl. Chem. 1980, 52, 2283–2302.
[14] Ch. Reichardt, Solvents and Solvent Effects in Organic Chemistry, VCH,
Weinheim, 1990.
Kinetic procedure
´
[15] E. Sledziewska, K. Bogdanowicz-Szwed, Reguły Zachowania Symetrii
´
Orbitali w Reakcjach Pericyklicznych, UJ, Krakow, 1980.
´
The rates of the cycloaddition reactions were followed by
integrating the area (At) of the HPLC peaks corresponding to the
cycloadducts 3 and 4. The analyses were carried out at 25 8C.
At this temperature, the cycloadducts 3 and 4 obtained from
each pair of reactants had the same Rt. The starting reaction
mixtures were prepared by introducing a weighed quantity of an
appropriate E-2-arylnitroethene to the solution of cyclopenta-
diene in a suitable solvent. The initial concentration of the latter
reactant was ca. 0.005 mol/l, whereas the former one was always
used in 20–22-fold molar excess. The reaction mixtures were
placed in ampoules of 1 ml capacity and maintained at constant
temperatures (Table 4). During the kinetic runs, the ampoules
were taken periodically from the thermostat and quickly cooled
´
[16] R. Jasinski, M. Kwiatkowska, A. Baranski, J. Mol. Struct. (TheoChem)
2009, 910, 80–87.
´
[17] R. Jasinski, A. Baranski, J. Mol. Struct. (TheoChem) 2010, 949, 8–13.
[18] M. Kwiatkowska, R. Jasinski, M. Mikulska, A. Baranski, Monatsh. Chem.
´
´
´
2010, 141, 545–548.
´
´
[19] R. Jasinski, M. Rzyman, A. Baranski, Coll. Czech. Chem. Commun. 2010,
75, 919–929.
[20] L. R. Domingo, M. J. Aurell, P. Perez, R. Contreras, Tetrahedron 2002,
58, 4417–4423.
[21] J. M. Gonzales, C. J. Barden, S. T. Brown, P. R. Schleyer, H. F. Schaefer,
Q. S. Li, J. Am. Chem. Soc. 2003, 125, 1064–1071.
´
[22] A. Baranski, E. Cholewka, Chem. Papers 1991, 45, 449–455.
[23] J. Cioslowski, A. Baranski, T. Juska, Tetrahedron 1986, 42, 4549–4555.
[24] V. D. Kiselev, A. G. Sakhabutdinov, I. M. Shakirov, V. V. Zverev, A. I.
Konovalov, Zh. Org. Khim. 1992, 28, 2244–2251.
wileyonlinelibrary.com/journal/poc
Copyright ß 2011 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 2011, 24 843–853

CYCLOADDITION REACTION OF CYCLOPENTADIENE
[25] J. Bourguignon, G. L. Nard, G. Queguiner, Can. J. Chem. 1985, 63,
2354–2361.
[26] W. E. Noland, B. A. Langager, J. W. Manthey, A. G. Zacchei, D. L. Petrak,
G. L. Eian, Can. J. Chem. 1967, 45, 2969–2978.
[27] S. S. Novikov, G. A. Shvekhgeimer, A. A. Dudinskaya, Izv. AN SSSR (Ser.
Khim) 1961, 690–695.
[28] G. I. Poos, J. Kleis, R. R. Wittekind, J. D. Rosenau, J. Org. Chem. 1961, 26,
4898–4904.
[37] J. G. Martin, K. Hill, Chem. Rev. 1961, 61, 537–562.
[38] D. Ranganathan, C. B. Rao, S. Ranganathan, A. K. Mehrotra, R. Iyengar,
J. Org. Chem. 1980, 45, 1185–1189.
[39] K. Schwetlick, Kinetische Metoden zur Untersuchung von Reaktions-
mechanismen, VEB, Berlin, 1971.
[40] R. G. Parr, W. Yang, Density Functional Theory of Atoms and Molecules,
Oxford University, New York, 1989.
[41] I. Benghiat, E. I. Becker, J. Org. Chem. 1958, 23, 885–890.
[42] R. Sustmann, Tetrahedron Lett. 1971, 2721–2724.
[43] V. D. Kiselev, A. I. Konovalov, J. Phys. Org. Chem. 2009, 22,
466–483.
[29] E. Pretsch, P. Bullmann, C. Affolter, Structure Determination of Organic
Compounds; Tables of Spectral Data, Springer-Verlag, Berlin, Heidel-
berg, 2000.
[30] J. L. Holmes, D. Mc Gillivaray, Org. Mass Spectrom. 1971, 5, 1339–1341.
[31] H. Kwart, T. A. Blazer, J. Org. Chem. 1970, 35, 2726–2731.
[32] T. Goto, A. Tatematsu, Y. Hata, R. Muneyuki, H. Tanida, K. Tori,
Tetrahedron 1966, 22, 2213–2222.
[33] N. S. Vul’fson, V. G. Zaikin, A. I. Mikaya, Mass-spektrometrya Organi-
cheskikh Soedinenii, Izd. Khimya, Moscow, 1986.
[34] H. Egsgaard, in The Chemistry of Amino, Nitroso, Nitro and Related
Groups (Ed.: S. Patay), J. Wiley & Sons, New York, 1996.
[35] J. T. Larkins, F. E. Saalfeld, L. Kaplan, Org. Mass Spectrom. 1969, 2,
213–221.
[44] W. L. Jorgensen, D. Lim, J. F. Blake, J. Am. Chem. Soc. 1993, 115,
2936–2942.
[45] G. S. Hammond, J. Am. Chem. Soc. 1955, 77, 334–338.
[46] G. Leroy, M. Sana, L. A. Burke, M. T. Nguyen, Quantum Theory Chem.
React. 1980, 1, 91–144.
[47] R. T. Arnold, G. P. Scott, Org. Synth. Coll. 1963, 4, 238–241.
[48] M. J. Kamlet, J. Am. Chem. Soc. 1955, 77, 4896–4898.
[49] S. Byrdy, Z. Eckstein, J. Plenkiewicz, Bull. Acad. Pol. Sci. (Chem.) 1961, 9,
627–631.
[50] N. Campbell, W. Anderson, J. Gilmore, J. Chem. Soc. 1940, 446–451.
[51] A. Leniewski, in Preparatyka i Elementy Syntezy Organicznej (Ed.: A.
[36] V. V. Razin, in Sovemiennye Problemy Organicheskoi Khimii, Izd LGU,
Leningrad, 1989.
´
Wrobel), PWN, Warszawa, 1993.
J. Phys. Org. Chem. 2011, 24 843–853
Copyright ß 2011 John Wiley & Sons, Ltd.
wileyonlinelibrary.com/journal/poc