Frontier orbital interactions in the hetero Diels-Alder cycloaddition of diazadienes

Article abstract of DOI:10.1139/V08-035

This work deals with the molecular orbital calculation studies performed on different diazadienes to assess their reactivity pattern. The interaction of these diazadienes with various electron-poor and electron-rich dienophiles leads to the formation of diazines and tetrazines as the cycloadducts. The results from frontier orbital interactions were used to rationalize the reactivity and predictability of NDAC and IEDDAC reaction pathways. Correlation studies were also performed to predict reactivity sequence using a number of electronic descriptors, such as electrophilicity index (ω), chemical potential (μ), electronic charge ΔN_max, and chemical hardness η. Moreover, these studies exhibit good compatibility with experimental observations.

Full text of DOI:10.1139/V08-035

384
Frontier orbital interactions in the hetero
Diels–Alder cycloaddition of diazadienes
Pratibha Sharma, Ashok Kumar, Vinita Sahu, and Jitendra Singh
Abstract: This work deals with the molecular orbital calculation studies performed on different diazadienes to assess
their reactivity pattern. The interaction of these diazadienes with various electron-poor and electron-rich dienophiles
leads to the formation of diazines and tetrazines as the cycloadducts. The results from frontier orbital interactions were
used to rationalize the reactivity and predictability of NDAC and IEDDAC reaction pathways. Correlation studies were
also performed to predict reactivity sequence using a number of electronic descriptors, such as electrophilicity index
(ω), chemical potential (µ), electronic charge ∆Nmax, and chemical hardness η. Moreover, these studies exhibit good
compatibility with experimental observations.
Key words: AM1, MNDO, PM3, diazadienes, tetrazines, electrophilicity index, chemical potential.
Résumé : Dans ce travail, on rapporte les résultats obtenus lors de calculs d’orbitales moléculaires effectués sur divers
diazadiènes dans le but d’évaluer leur mode de réactivité. L’interaction de ces diazadiènes avec divers diénophiles ri-
ches et pauvres en électrons conduit, suivant le cas, à la formation de diazines et de tétrazines comme cycloadduits. On
a utilisé les résultats d’interactions d’orbitales frontières pour rationaliser la réactivité et pour vérifier si les voies réac-
tionnelles de cycloaddition de Diels–Alder normale et de cycloaddition de Diels–Alder à demande inversée d’électrons
peuvent être utilisées pour les prédire. Utilisant un certain nombre de descripteurs électroniques, tel l’indice du carac-
tère électrophile (ω), le potentiel chimique (µ), la charge électronique (∆Nmax) et la dureté chimique (η), on a aussi ef-
fectué des études de corrélation pour prédire la séquence de réactivité. De plus, ces études présentent une bonne
compatibilité avec les observations expérimentales.
Mots-clés : modèle Austin-1 (“AM1"), négligence modifiée du recouvrement différentiel (”MNDO"), modèle paramé-
trisé numéro 3 (“PM3"), diazadiènes, tétrazines, indice du caractère électrophile, potentiel chimique.
[Traduit par la Rédaction] Sharma et al. 394
Introduction
predict selectivity and reactivity of compounds. A system-
atic perusal of the literature (5) reveals that the prediction
and reactivity of DA reaction are normally based on the
strength of HOMO–LUMO interactions between diene and
dienophlie. In view of these precedents and as part of our re-
search program devoted to the study of heterocycles (6), we
have made an effort to interpret experimental outcomes through
normal Diels–Alder cycloaddition (NDAC) and inverse
electron-demand Diels–Alder cycloaddition (IEDDAC)
pathways. Moreover, we have also incorporated electronic
descriptors, such as the electrophilicity index ω, the chemi-
cal potential µ, and ∆Nmax, to interpret cycloaddition reactivity.
A number of diazadienes bearing one or more electron-
withdrawing and electron-donating groups have been chosen
to initialize these studies using semi-empirical probes (7).
These dienes undergo DA reactions with a variety of
dienophiles to achieve tetrazines as the final product in ap-
preciable yield under optimum reaction conditions. A judi-
cious selection of the best method to predict the FMO
energies of dienes and various dienophiles has been done. A
comparative evaluation of different Hamiltonians for various
dienes is shown in Table 1, and the AM1 method was found
to show good correlation with Fukui’s approximations.
Hence, we have chosen the AM1 method for further discus-
sion. Moreover, it is to be emphasized here that the MO
semi-empirical method allows the study of more complex
diazadienes in a very much more facile manner in compari-
The Diels–Alder (1) (DA) reaction is one of the most
prominent synthetic strategies in organic chemistry to craft
annular compounds from small fragments. The DA reactions
of 2-azadienes (2) is a very useful method for providing
rapid access to a range of highly substituted biologically
active heterocyclic system like pyridines, dihydropyridines,
and tetrazine (3). The development of synthetic methodolo-
gies based on aza-DA reactions to obtain six-membered
tetrazines has attracted much interest because of their utility
in the field of medicinal chemistry as antitumor drugs and in
energetic chemistry as powerful explosives. Hence, the DA
reaction is found to be most a promising method for
tetrazine synthesis. With the advancement of computational
chemistry to predict the fate of a reaction, herein we have
tried to incorporate the quantum mechanical results to inter-
pret DA cycloaddition. In this context, the Fukui’s FMO
method (4) is found to be the most suitable synthetic tool to
Received 29 November 2007. Accepted 19 February 2008.
Published on the NRC Research Press Web site at
canjchem.nrc.ca on 27 March 2008.
P. Sharma,1 A. Kumar, V. Sahu, and J. Singh. School of
Chemical Sciences, Devi Ahilya University, Takshashila
Campus, Indore, MP 452 001, India.
1Corresponding author (e-mail: drpratibhasharma@yahoo.com).
Can. J. Chem. 86: 384–394 (2008)
doi:10.1139/V08-035
© 2008 NRC Canada

Sharma et al.
385
Table 1. Frontier orbital energies of diazadiene.
AM1
MNDO
HOMO
PM3
Dienes
HOMO
LUMO
LUMO
HOMO
LUMO
0.465
0.135
0.822
–0.245
–0.036
–1.353
–0.126
–0.237
–0.856
–0.091
0.282
–0.011
0.120
–0.231
–0.036
–0.915
–0.123
–0.273
–0.327
–0.091
–9.180
–9.677
–10.157
–9.166
–8.544
–9.800
–8.486
–8.864
–9.397
–9.853
0.388
0.106
0.147
–0.552
–0.041
–1.326
–0.501
–0.529
–0.986
–0.574
1
2
3
4
5
6
7
8
–9.356
–9.903
–9.998
–9.314
–8.319
–9.794
–8.407
–8.875
–9.456
–9.352
–9.502
–9.982
–9.691
–9.310
–9.358
–9.668
–8.812
–8.899
–9.526
–9.440
9
10
Fig. 1. Frontier orbital energies for 1,3-butadiene 1, 2-aza-diene 2, and diazadiene 3 and 4 (AM1 calculation).
son to ab initio methods. To make a comparative study, we
have used some reference dienes and have revealed that 1,3-
butadiene (1), being an electron-rich diene is prone to un-
dergo NDAC reaction with electron-deficient dienophlie.
However, the similar prediction is not easy for heterodienes,
because upon substituting one CH moiety of 1,3-butadiene
with N, the energy of 2 HOMO–LUMO decreases, while
substitution with another N does not show a major decrease
in HOMO energy of diene 3 (Figs. 1 and 2).
Furthermore, upon comparing the relative reactivity of
diazadienes 4–10, it has been inferred that diazadiene 5,
bearing an electron-donating substituent (i.e., –NMe2) at the
phenyl ring, was expected to be more electron-rich com-
pared with 4, and therefore it is more likely to react with
electron-deficient dienophile through the NDAC pathway.
On the other hand, 6 takes part in the IEDDAC reaction only
with electron-rich dienophile. Similarly, diazadienes 7 and
8, bearing electron-donating substituents, also show the pre-
cedence of the NDAC pathway over IEDDAC with electron-
deficient dienophile.
The CH÷N substitution in the dienes enhances their reac-
tivity towards electron-rich dienophiles in IEDDAC reac-
tions. However, further decreases in electrophilicity by
electron-withdrawing substituent, as in 9 and 6, makes them
more favored species for IEDDAC reactions.
Also, in an attempt to synthesize tetrazines, reference
diene 4 was allowed to react with a number of dienophiles
11–20 (Figs. 3 and 4) and the experimental results were cor-
related with theoretical outcomes. The AM1 optimized ge-
ometry of cycloadduct 25 is depicted in Fig. 5.
© 2008 NRC Canada

386
Can. J. Chem. Vol. 86, 2008
Fig. 2. HOMO and LUMO pictures of dienes 1–3.
assumption into consideration and tried to associate the
experimental results with computational semi-empirical out-
comes. Hence to gain deeper insight into the DA cyclo-
addition strategy, we proceeded with the treatment of
diazadiene 4 with different dienophiles 11–20 and the perti-
nent results are presented in Tables 2 and 3.
Here we offer an interesting and novel interpretation of
eigenfunctions to predict the outcomes of DA reactions. In
this context, diazadiene 4, which is an electron-rich diene,
shows similar reactivity with maleic anhydride 16, bis-4-
nitrophenyl diazene 19, and 4-[4-nitrophenyl]-diazenyl phe-
nol 20, which are electron-deficient dienophiles. The reac-
tion time required for 4 and dienophiles 16, 19, and 20 was
found to be approximately similar, favoring NDAC pathway.
Moving toward more electron-rich dienophiles, other than
16, 19, and 20, the reaction time increases and sometimes no
reaction occurs. 1,2-Ethoxyethene 11 and allyl chloride 18
should take the longest time for reaction with 4. In actuality,
the reactions of 4 with allyl halide 18 requires 13 h, which
may be due to the high HOMO–LUMO energy gap for this
NDAC reaction, which includes a significant molecular or-
bital energy gap. A close inspection of eigenfunctions sug-
gests that the dienophile 1,2-ethoxyethene 11, N,N-dimethy-
4-phenyldiazenyl aniline 12, and 4-phenyldiazenyl aniline
13, are likely to undergo IEDDAC reaction, and the experi-
mental results show that dienophiles 11 and 12, being highly
electron-rich dienophiles, remain unreactive with diene 4 af-
ter a long time. According to the energy level gap between
participating molecular orbitals for dienophiles 12 and 13,
both are prone to undergo the IEDDAC pathway, but experi-
mentally no reaction occurs with diazadiene 4 under similar
reaction conditions. With dienophile 4-phenyldiazenyl phe-
nol 14, 1-(dinitrophenyl)-2-phenyldiazene 15, and diethyl
fumarate 17, reactions proceed according to our expecta-
tions, and the experimental results are in agreement with
computational outcomes with nearly similar reactivity. It has
been experimentally found that the normal DA cycloaddition
of diazadiene 4 with allyl chloride 18 provides better yield
of the adduct, although the HOMO–LUMO energy gap is
larger for these reactions, in contrast with other cyclo-
additions.
Reactivity of diazadienes 4–10
Table 1 shows the frontier orbital energies of diazadienes
4–10 calculated at the AM1 level. The AM1 calculations
described above were performed considering the S-cis con-
formations of dienes.
It is deduced from the aforementioned discussion that
HOMO–LUMO energies lowers upon CH÷N substitution,
making them more reactive species towards electron-rich
dienophiles. Substitution of an electron-donating group
(–NMe2) at the phenyl ring raises the HOMO energy by
0.995 eV and lowers the LUMO energy, thereby explaining
their preferred participation in NDAC reactions. The calcu-
lated HOMO energies of 4, 6, 9, and 10 are lower than those
of 5, 7, and 8. This explains the lack of reactivity of these
diazadienes towards electron-deficient dienophiles (NDAC).
On the other hand, the predicted LUMO energies of 4, 6–10
are lower than that of 5, pointing to an increased reactivity
toward electron-rich dienophiles (IEDDAC). Diazadienes 6,
bearing a nitrophenyl group, has the lowest energy followed
by 9, 10, and 4. This indicates that the addition of one nitro
group at the 4-position of phenyl ring of diazadienes has a
more substantial effect than the introduction of N at the
same position.
Discussion
A perusal of the calculated energies (Table 2) of the fron-
tier orbitals for the diazadiene 4 and the dienophile points
out the significance of the HOMO–LUMO energy gap
(Fig. 6). Here Lde-Hdo represents the calculated energy gap
(eV) between each dienophile with respect to the LUMO of
diene 4. Ldo-Hde represents the calculated energy gap (eV)
between the dienophile’s LUMO with respect to the HOMO
of diene 4. The Ldo-Hde energy gap is least (7.695 eV) for
diazadiene 4 and maleic anhydride 16. As 16 is an electron-
deficient dienophile, this energy gap favors the NDAC reac-
tion, while the highest energy gap for 1,2-ethoxyethene 11
makes this reaction IEDDAC. Furthermore, it is interesting
to note that the energy gap (Lde-Hdo) between diazadiene 4
and N,N-dimethy-4-phenyldiazenyl aniline 12, and 4-
phenyldiazenyl aniline 13 and 4-phenyldiazenyl phenol 14
dienophiles, are nearly similar, but their reactivities differ
markedly, which further emphasizes that the Lde-Hdo en-
It can be concluded that on the basis of HOMO energy
from AM1 calculations, the reactivity of different diazadienes
to participate in inverse electron-demand DA reactions
would be in the following order: 6 > 9 > 10 > 4 > 8 > 7 > 5.
NDAC and IEDDAC
In a view to unravelling the understanding of the mecha-
nistic details of DA cycloaddition, we have taken the Fukui’s
© 2008 NRC Canada

Sharma et al.
387
Fig. 3. Orbital energies (eV) and geometries of dienophile 11–20.
ergy gap is not a reliable tool to predict the reactivity of
dienes and dienophiles toward IEDDAC reactions. However,
the Ldo-Hde energy gap reasonably predicts the reactivity of
dienes and dienophiles. Hence, the NDAC is a more favored
pathway for DA reactions than is IEDDAC.
the vertical ionization potential (I) and electron affinity (A)
using Koopman’s theorem. Here, in our calculations I =
EHOMO and A = ELUMO were used.
The electrophilicity index encompasses both the propen-
sity of the electrophile to acquire an additional electronic
charge driven by µ2 (the square of the electronegativity) and
the resistance of the system to exchange electronic charge
with the environment described by η simultaneously. The
maximum electronic charge Nmax is another useful quantity;
it is the maximal electronic charge which electron accept
from environment, giving a better understanding of Fukui’s
model.
Table 4 presents an overview of electronic parameters
such as electronic chemical potential µ, chemical hardness η,
and global electrophilicity index ω. The electronic chemical
potential for dienophile 11–20 lies in the range –6.821 to
–3.664, whereas for diene 4 this value is an intermediate
one, –4.779. Dienophiles11, 12, and 13 have higher chemi-
cal potentials than 4, which further suggests that a net
charge transfer will take place from electron-rich dienophiles
to electron-poor diene 4 in an IEDDAC reaction pathway. In
addition, a good correlation can be observed between the
substitution pattern on dienophile and the chemical poten-
tial µ for the remaining dienophiles 14–20, where charge
transfer take place from an electron-rich diene to an
electron-deficient dienophile through the NDAC pathway.
Global electrophilicity index
For a long time Domingo’s group has been interested in
the study of the molecular mechanism of the polar DA reac-
tions (8–9). Recently, the use of the global electrophilicity
index ω proposed by Parr et al. (10), has been reported to
classify the global electrophilicity of a series of dienes and
dienophiles currently present in DA reactions (11). We
found a good correlation between the difference in electro-
philicity for the diene and dienophile pair, ∆ω, and the
feasibility of the cycloaddition. Therefore, ∆ω for a diene–
dienophile pair is a valuable tool to predict the polar charac-
ter of a DA reaction. The global electrophilicity index ω,
which measures the stabilization in energy when the system
acquires an additional electronic charge Nmax from the envi-
ronment, has been given the following simple expression,
ω = µ2/2 η
where µ ≈ (I + A)/2 and η = (A – I) are the electronic chemi-
cal potential and chemical hardness of the ground state of at-
oms and molecules, respectively, approximated in terms of
© 2008 NRC Canada

388
Can. J. Chem. Vol. 86, 2008
Fig. 4. Various cycloadducts 21–30.
Fig. 5. Optimized geometry of tetrazine cycloadduct 25.
energy gap (7.695) in DA cycloaddition. Experimentally this
reaction took the least reaction time to completion (-2 h) in
the presence of xylene.
The highest µ value of –3.664 for 11 opens up the possi-
bility of an inverse electron-demand DA reaction in which
charge transfer direction is from an electron-rich dienophile
to an electron-deficient diene, making this reaction experi-
mentally not feasible under similar reaction conditions. The
electrophilicity index ω values for the diene and dienophile
falls within the range 18.08–60.84 eV; the highest ω value
obtain for 16 illustrates the highly electrophilic nature of this
dienophile and is supported by experimental observations.
Moreover, the difference in electrophilicity between the
diazadiene 4 and dienophiles (∆ω) gives a measure of the re-
activity of diene and dienophiles in a cycloaddition reaction,
where the smallest value of ∆ω for dienophile 16 suggests
the nonpolar synchronous nature of transition states in nor-
mal DA cycloaddition.
On the other hand, the highest ∆ω difference value for the
4a+1 reaction provides an asynchronous polar transition
state and IEDDAC pathway, which is experimentally not ob-
servable and no reaction occurs at all.
The µ value is found to be lowest for 16 amongst dienohiles
11–20, indicating ease of reaction with the least Ldo-Hde
© 2008 NRC Canada

Sharma et al.
389
Table 2. Calculated HOMO and LUMO energies of diazadiene 4 and dienophiles 11–20.
S. No.
Diene–dienophile
HOMO
LUMO
Ldo-Hde
Lde-Hdo
4
11
12
13
14
15
16
17
18
19
20
–9.314
–8.713
–8.433
–8.667
–9.215
–8.952
–12.023
–11.316
–10.473
–10.578
–9.648
9.069
10.699
9.098
9.049
8.893
8.136
7.695
8.272
9.983
7.735
7.999
9.069
8.468
8.188
8.422
8.970
1
2
3
4
5
6
7
8
–0.245
1.385
–0.216
–0.265
–0.421
–1.178
–1.619
–1.042
0.669
8.707
11.778
11.071
10.228
10.333
9.403
9
10
11
–1.579
–1.315
Table 3. Experimental results of the [4+2] cycloaddition of diene 4 with dienophiles 11–20.
Entry
Dienophile
Adduct
Time to completion (h)
Yield (%)
1
2
3
4
5
6
7
8
9
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
—
—
—
20
8
—
—
—
65
70
85
75
70
79
78
2
6.5
13
4
10
6
Fig. 6. Molecular orbital energy gap between diene 4 and dienophile 11 and 16.
Multiple analysis regression (MAR) data between the
electrophilicity index ω and electron affinity EA for 11 com-
pounds (Table 4) shows that the statistical parameters (stan-
dard deviation) SD ≈ 0.31 and (correlation co-efficient) R ≈
0.95 significantly support excellent correlation (Fig. 7) be-
tween the electrophilicity index ω and electron affinity EA.
The stereo- and regio-selectivity of DA cycloadditions are
easily rationalized by examining the orbital coefficient val-
ues on the molecular orbitals. The preferred regioisomeric
transition states in DA cycloadditions depends upon the ter-
minal coefficient of the interacting orbitals. Table 5 shows
the orbital coefficient at different carbon atoms of dienes 1–
© 2008 NRC Canada

390
Can. J. Chem. Vol. 86, 2008
Table 4. Global electronic descriptors values of diene and dienophiles.
Entry
Diene/dienophile
HOMO
LUMO
µ (a.u.)
η (a.u.)
ω (eV)
Nmax
ω (eV)
1
2
3
4
5
6
7
8
9
4
11
12
13
14
15
16
17
18
19
20
–9.314
–8.713
–8.433
–8.667
–9.215
–8.952
–12.023
–11.316
–10.473
–10.578
–9.648
9.069
10.098
8.217
8.402
8.794
0.00
–0.245
+1.385
–0.216
–0.265
–0.421
–1.178
–1.619
–1.042
+0.669
–1.579
–1.315
–4.779
–3.664
–4.324
–4.466
–4.818
–5.065
–6.821
–6.179
–4.902
–6.078
–5.482
34.26
18.08
30.95
32.30
35.91
44.90
60.84
50.56
29.34
55.86
49.05
0.527
0.363
0.526
0.532
0.548
0.652
0.656
0.601
0.440
0.675
0.660
+16.180
+3.310
+1.960
–1.650
–10.640
–26.580
–16.300
+4.920
–21.600
–14.790
7.774
10.404
10.274
11.142
8.999
10
11
8.333
Fig. 7. Correlation graph vs. electrophilicity index ω and electron
affinity EA.
withdrawing and favorable for NDAC reactions. 1-
(Dinitrophenyl)-2-phenyldiazene 15 and 4-[4-nitrophenyl]-
diazenyl phenol 20, having at least one electron-withdrawing
moiety attached to the phenyl ring, show the low orbital co-
efficient at N-2, which is in accordance with literature data.
Similarly, 4-phenyldiazenyl aniline 13 and 4-phenyldiazenyl
phenol 14, having electron-releasing substiutents at the
phenyl ring, possess larger orbital coefficients at N-1 than at
N-2, making these species regiospecific towards unsymmet-
rical dienes. However, similar predictions cannot be made
for N,N-dimethy-4-phenyldiazenyl aniline 12, which remains
unreactive in DA cycloaddition with diazadiene 4.
Regioselective studies
When an unsymmetrical diene 10 is allowed to react with
a certain range of the aforementioned unsymmetrical dieno-
philes, the regioselectivity of the cycloadduct was also taken
into consideration. We have calculated atomic orbital coeffi-
cients at interacting orbitals to check their participation in
DA cycloaddition. The orbital coefficient values on diaza-
dienes are very much lower than for their carbon congeners.
A close inspection of the coefficients of diene 10 and dieno-
phlie 15 suggest the preferred formation of regioselective
R-31b over R-31a. The experimental results were in accor-
dance with these findings. For a better understanding of
regioselectivty in hetero DA reactionsm, diene 10 was also
treated with dienophile 20 and preferred formation of R-32a
rather than R-32b was in support of Houk’s assumption.
Both of these reactions follow the NDAC pathway under
similar reaction conditions (Fig. 8).
Furthermore, the role of secondary orbital interactions in
deciding the regioselectivity of the product was also envis-
aged in the case of unsymmetrical diene 10. It has been re-
vealed that when diazadiene 10 was allowed to react with
dienophile 15, R-31b was solely obtained as the regio-
selective cycloadduct because of primary orbital interactions
between terminal orbital coefficients of the HOMO diene
and the LUMO dienophile. It has been noticed that preferred
regioisomers can still be predicted by considering secondary
orbital interactions between the N-2 and N-3 positions of
diene 10 and the C6H4-NO2 moiety of dienophlie 15. A
close inspection of the orbital coefficients at N-2 and N-3 in
diene 10 show that N-2 has higher orbital coefficient than N-
3. Thus, the stabilization of an endo transition state is
greater when the C6H4-NO2 group of dienophile 15 is near
4 and 10. The results demonstrate that in 1,3-butadiene 1,
the terminal coefficient at C-1 and C-4 of HOMO wave
functions are of equal magnitude, but upon CH→N substitu-
tion the coefficient magnitude suddenly decreases and so 1-
azadienes becomes regiospecific for the unsymmetrical sub-
stituted dienophile. Introduction of N in 1,3-butadiene en-
hances the electrophilicity of the diene, making it a preferred
species for IEDDAC reaction. In azadienes, the C-1 position
is more polarized than the C-4 position. A similar decrease
in orbital coefficient is found upon introducing another N in
the diene moiety. A close examination of the orbital coeffi-
cients on azadienes 2 and 3 reveals that the magnitude of the
orbital coefficients is larger on the C-3/N-3 position than on
the C-2/N-2 position, which could play an immense role in
the preferred formation of regioisomers through secondary
orbital interactions.
An exhaustive exploration of orbital coefficients on differ-
ent dienophiles (Table 6) provides regioselective understand-
ing toward the DA cycloadduct. The orbital coefficients of
the symmetrical molecule (dienophile) are identical in mag-
nitude at terminal atomic centers. These studies show that
D-1 (here D-1 is taken as reference dienophile to interpret
the comparative results of dienophiles 11–20 in terms of or-
bital coefficients, C6H5-N=N-C6H5) has a similar orbital co-
efficient at N-1 and N-2 in LUMO, while these values are
considerably lower due to the electron-withdrawing phenyl
substituent, which makes this species more electron-
© 2008 NRC Canada

Sharma et al.
391
Table 5. Orbital coefficient values at different diazadienes.
Orbital coefficient (HOMO)
Orbital coefficient (LUMO)
Diene
C-1
C-2/N-2
C-3/N-3
C-4
C-1
C-2/N-2
C-3/N-3
C-4
1
2
3
4
0.563
0.328
0.306
0.230
0.028
0.427
0.219
0.118
0.243
0.065
0.427
0.406
0.391
0.241
0.038
0.563
0.550
0.199
0.044
0.069
0.566
0.579
0.511
0.333
0.030
0.423
0.439
0.4354
0.395
0.081
0.423
0.368
0.012
0.030
0.041
0.566
0.579
0.021
0.056
0.071
10
exhibited high reactivity towards 15, which is a electron-
deficient species in NDAC pathway. (iii) Synthesis of differ-
ent substituted tetrazines were achieved via NDAC and
IEDDAC DA pathways. (iv) All the results were interpreted
in terms of electrophilicity index ω and a good correlation
was observed between global electronic parameters and ex-
perimental findings. Hence, the FMO interactions are the
key players to decide the fate of DA reactions. Similar stud-
ies on hetero DA reactions are in progress.
Table 6. Orbital coefficient values at different dienophiles.
Acknowledgements
Orbital coefficient
(HOMO)
Orbital coefficient
(LUMO)
We are grateful to the Central Drug Research Institute,
Lucknow, India for providing spectroanalytical facilities.
One of the authors (VS) is also thankful to the Council of
Scientific and Industrial Research (CSIR), India, for provid-
ing a junior research fellowship.
Dienophile
N-1
N-2
N-1
N-2
D-1
12
13
14
15
20
0.119
0.164
0.189
0.146
0.125
0.059
0.102
0.056
0.068
0.115
0.114
0.317
0.388
0.372
0.403
0.403
0.071
0.366
0.370
0.293
0.332
0.327
0.024
0.278
Experimental
Computational method
The AMl, MNDO, and PM3 approximations to molecular
orbital theory have been employed using the MOPAC 2007
computer program (12). The geometries were fully opti-
mized at minimum gradient level (0.01). The minimum en-
ergy conformations were used to compute molecular orbital
and charge distributions. To explain the results in terms of
global electronic parameters viz., electrophlicity index ω,
chemical potential µ, etc., the calculations were done at
MOPAC 2007 program.
the secondary position of the diene, yielding the R-31b iso-
mer as the favourable product.
Upon similar analysis, the regioselective formation of R-
32a can be explained by considering the secondary orbital
interactions between orbital coefficients at N-2 of diene 10
and the C6H4-NO2 moiety of dienophile 20. Thus, secondary
orbital interactions exerted a significant effect on the
regioselectivity of the Diels Alder reaction between unsym-
metrically substituted diene and dienophile.
General
All the chemicals used were of AR grade purity. IR spec-
tra were recorded on PerkinElmer model 377 spectro-
photometer in KBr pellets. H NMR spectra were recorded
on a Bruker DRX300 instrument. The FAB mass spectra
were recorded on a JEOLSX102/DA–6000 Mass Spectrome-
ter using argon–xenon (6 kv, 10 mA) as the FAB gas.
Analytical thin layer chromatography was performed using
E. Merck silica gel G (0.50 mm plates, Merck No. 5700).
The melting points were determined on an electric melting
point apparatus in open capillaries and are uncorrected.
Conclusion
1
Concluding, we have performed and interpreted FMO in-
teractions with a critical evaluation of the reaction pathways.
Highlights of all the studies can be summarized as follows.
(i) The relative reactivity of different diazadienes in DA re-
actions was studied in a selective and predictive way.
(ii) The Ldo-Hde energy gap was found to a be a favored
tool to predict DA cycloaddition. These predictions are in
good agreement with experimental findings where diene 4
© 2008 NRC Canada

392
Can. J. Chem. Vol. 86, 2008
Fig. 8. Predictions of regioselectivity for diazadiene 10 and dienophiles 15 and 20.
vent were distilled off at reduced pressure from the reaction
mixture and the crude cycloadduct was recrystallised and
purified by TLC resolution studies on silica gel (E Merck)
using ethyl acetate–xylene (4:6, v/v).
Synthesis of diazadiene 4
The diazadiene 4 was prepared by condensing together
dissolving hydrazine hydrate (0.05 mol) into 50% glacial
acetic acid (10 mL), and to this solution bezaldehyde
(0.10 mol) was added to obtained diazadiene 4 in good yield
(80%) as a yellow-green crystalline product. It was washed
with cold water and recrystallized from ethanol.
Adduct 24
The reaction was performed according to the general pro-
cedure starting with 1.98 gm of 14. The cycloadduct was
isolated in 65% yield, mp 112 °C, decomposed. IR
(KBr, cm–1) 3570 (–OH str.), 3033 (=C-H, sp2), 2986 (C-H,
sp3), 1621 (C=C), 1541 (N=N), 1458, 1371 (C-H, bending,
sp3), 1071 (C-O), 1046 (C-N), 886, 763, 676 (sub. phenyl).
1H NMR (ppm) δ: 3.15 (s, 1H, -CH-Ph), 3.37 (s, 1H, -CH-
Typical procedure for Diels–Alder reactions
To a solution of diazadiene 4 (0.01 mol) in dry xylene was
added dienophile 11–20 (0.01 mol) in equimolar quantity.
The solution was then refluxed for the required time (Ta-
ble 3) with constant stirring. At the end of the reaction, sol-
© 2008 NRC Canada

Sharma et al.
393
Ph), 6.85 (dd, 2H, -C6H4-OH, J = 8.5 Hz, J = 2.7 Hz), 7.17
(dd, 2H, –C6H4-OH, J = 8.1 Hz, J = 2.8 Hz), 7.21–7.67 (m,
15H, -phenyl), 11.75 (s, 1H, -OH). FAB-MS m/z: 407. Anal.
calcd. for C26H22N4O: C 76.83, H 5.46, N 13.78; found: C
76.78, H 5.41, N 13.72.
1535 (–NO2), 1451, 1361 (C-H, bending, sp3), 1041 (C-N),
881, 751, 672 (sub. phenyl). H NMR (ppm) δ: 3.16 (s, 2H,
–CH-Ph × 2), 7.44–7.66 (m, 10H, –phenyl), 7.70 (dd,
4H, –C6H4-NO2, J = 8.5 Hz, J = 2.3 Hz), 8.14 (dd, 4H,
–C6H4-NO2, J = 8.8 Hz, J = 2.5 Hz). FAB-MS m/z: 481.
Anal. calcd. for C26H20N6O4: C 64.99, H 4.20, N 17.49;
found: C 64.91, H 4.18, N 17.45.
1
Adduct 25
The reaction was performed according to the general pro-
cedure starting with 2.27 gm of 15. The cycloadduct was
isolated in 70% yield, mp 121–122 °C. IR (KBr, cm–1): 3034
(=C-H, sp2), 2974 (C-H, sp3), 1623 (C=C), 1544 (N=N),
1535 (–NO2), 1451, 1361 (C-H, bending, sp3), 1064 (C-O),
Adduct 30
The reaction was performed according to the general pro-
cedure starting with 2.43 gm of 20. The cycloadduct was
isolated in 78% yield, mp 128–130 °C. IR (KBr, cm–1): 3556
(–OH str.), 3028 (=C-H, sp2), 2987 (C-H, sp3), 1615 (C=C),
1548 (N=N), 1548 (–NO2), 1467, 1372 (C-H, bending, sp3),
1
1041 (C-N), 881, 751, 672 (sub. phenyl). H NMR (ppm) δ:
3.11 (s, 1H, –CH-Ph), 3.22 (s, 1H, –CH-Ph), 7.45–7.64 (m,
15H, -phenyl), 7.74 (dd, 2H, –C6H4-NO2, J = 8.6 Hz, J =
3.0 Hz), 8.11 (dd, 2H, –C6H4-NO2, J = 9.1 Hz, J = 2.2 Hz).
FAB-MS m/z: 436. Anal. calcd. for C26H21N5O2: C 71.71, H
4.86, N 16.08; found: C 71.67, H 4.79, N 16.01.
1
1061 (C-O), 1044(C-N), 884, 761, 669 (sub. phenyl). H
NMR (ppm) δ: 3.19 (s, 1H, –CH-Ph), 3.32 (s, 1H, –CH-Ph),
6.78 (dd, 2H, –C6H4-OH, J = 7.9 Hz, J = 2.9 Hz), 7.31 (dd,
2H, –C6H4-OH, J = 8.1 Hz, J = 3.0 Hz), 7.40–7.63 (m, 10H,
–phenyl), 7.75 (dd, 2H, –C6H4-NO2, J = 8.5 Hz, J = 2.8 Hz),
7.91 (dd, 2H, –C6H4-NO2, J = 8.7 Hz, J = 2.9 Hz). FAB-MS
m/z: 452. Anal. calcd. for C26H21N5O3: C 69.17, H 4.69, N
15.51; found: C 69.11, H 4.65, N 15.49.
Adduct 26
The reaction was performed according to the general pro-
cedure starting with 0.98 gm of 16. The cycloadduct was
isolated in 85% yield, mp 124–125 °C. IR (KBr, cm–1): 3028
(=C-H, sp2), 2884 (C-H, sp3), 1778 (–C=O, anhydride), 1612
(C=C), 1544 (N=N), 1444, 1348 (C-H, bending, sp3), 1071
Adduct R-31b
The reaction was performed according to the general pro-
cedure starting with 2.27 gm of 15. The cycloadduct was
isolated in 71% yield, mp 110–112 °C. IR (KBr, cm–1): 3033
(=C-H, sp2), 2981 (C-H, sp3), 1618 (C=C), 1542 (N=N),
1539 (–NO2), 1441, 1371 (C-H, bending, sp3), 1061 (C-O),
1
(C-O), 1049 (C-N), 761, 688 (sub. phenyl). H NMR (ppm)
δ: 2.55 (d, 2H, –CH × 2, J = 5.8 Hz), 3.21 (d, 2H, –CH-Ph ×
2, J = 6.1 Hz), 6.83–7.81 (m, 10H, –phenyl). FAB-MS m/z:
307. Anal. calcd. for C18H14N2O3: C 70.58, H 4.61, N 9.51;
found: C 70.51, H 4.59, N 9.48.
1
1044 (C-N), 880, 749, 671 (sub. phenyl). H NMR (ppm) δ:
2.62 (d, 3H, –CH3), 2.68 (q, 1H, –CH-CH3), 3.23 (s, 1H,
–CH-C6H5), 6.91–7.43 (m, 10H, –phenyl), 7.51 (dd, 2H,
–C6H4-NO2, J = 8.8 Hz, J = 2.4 Hz), 7.84 (dd, 2H,
–C6H4-NO2, J = 8.9 Hz, J = 2.9 Hz). FAB-MS m/z: 375.
Anal. calcd. for C21H19N5O2: C 67.55, H 5.13, N 18.76;
found: C 67.14, H 5.08, N 18.61.
Adduct 27
The reaction was performed according to the general pro-
cedure starting with 1.44 gm of 17. The cycloadduct was
isolated in 75% yield, mp 130–132 °C. IR (KBr, cm–1): 3034
(=C-H, sp2), 2889 (C-H, sp3), 1771 (–C=O, ester), 1621
(C=C), 1551 (N=N), 1434, 1371 (C-H, bending, sp3), 1061
1
(C-O), 1053 (C-N), 775, 669 (sub. phenyl). H NMR (ppm)
Adduct R-32a
δ: 2.38 (d, 2H, –CH × 2, J = 5.5 Hz), 3.33 (d, 2H, –CH-Ph ×
2), 3.80 (s, 6H, –COOCH3), 6.81–7.66 (m, 10H, –phenyl).
FAB-MS m/z: 353. Anal. calcd. for C20H20N2O4: C 68.17, H
5.72, N 7.95; found: C 68.14, H 5.67, N 7.89.
The reaction was performed according to the general pro-
cedure starting with 2.43 gm of 20. The cycloadduct was
isolated as reddish brown solid in 75% yield, mp 135–
136 °C. IR (KBr, cm–1): 3549 (–OH str.), 3014 (=C-H, sp2),
2981 (C-H, sp3), 1614 (C=C), 1543 (N=N), 1551 (–NO2),
1457, 1371 (C-H, bending, sp3), 1059 (C-O), 1041 (C-N),
Adduct 28
The reaction was performed according to the general pro-
cedure starting with 0.76 gm of 18. The cycloadduct was
isolated in 70% yield, mp 78–80 °C. IR (KBr, cm–1): 3032
(=C-H, sp2), 2891(C-H, sp3), 1615 (C=C), 1557 (N=N),
1450, 1368 (C-H, bending, sp3), 550 (C-Cl), 1050 (C-N),
1
881, 758, 662 (sub. phenyl). H NMR (ppm) δ: 2.30 (d, 3H,
–CH3), 2.53 (s, 1H, –CH-CH3), 3.32 (q, 1H, –CH-C6H5),
6.54 (dd, 2H, –C6H4-OH, J = 7.9 Hz, J = 2.9 Hz), 6.79 (dd,
2H, –C6H4-OH, J = 8.1 Hz, J = 2.3 Hz), 6.99 (s, 5H,
–C6H5), 7.75 (dd, 2H, –C6H4-NO2, J = 8.5 Hz, J = 2.8 Hz),
7.92 (dd, 2H, –C6H4-NO2, J = 8.8 Hz, J = 2.7 Hz), 11.16 (s,
1H, –OH) ; FAB-MS m/z: 390. Anal. calcd. for C21H19N5O3:
C 64.77, H 4.92, N 17.98; found: C 64.61, H 4.85, N 17.89.
1
765, 661 (sub. phenyl). H NMR (ppm) δ: 2.48 (d, 2H,
–CH2–), 2.61 (d, 1H, –CH-C6H5, J = 5.9 Hz), 2.71 (m, 1H,
–CH-), 2.78 (d, 1H, –CH-C6H5, J = 5.7 Hz), 2.98 (d, 2H,
–CH2-Cl, J = 6.5 Hz), 7.22–7.58 (m, 10H, –phenyl).
FAB-MS m/z: 285. Anal. calcd for C17H17ClN2: C 71.70, H
6.02, N 9.84; found: C 71.67, H 5.92, N 9.78.
References
Adduct 29
1. (a) J. Sauer and R. Sustmann. Angew. Chem. Int. Ed. Engl. 19,
779 (1980); (b) K.N. Houk, Y.T. Lin, and F.K. Brown. J. Am.
Chem. Soc. 108, 554 (1986); (c) W.T. Borden, R.J.
Loncharich, and K.N. Houk. Annu. Rev. Phys. Chem. 39, 213
(1988).
The reaction was performed according to the general pro-
cedure starting with 2.72 gm of 19. The cycloadduct was
isolated in 79% yield, mp 79–80 °C. IR (KBr, cm–1): 3034
(=C-H, sp2), 2974 (C-H, sp3), 1623 (C=C), 1544 (N=N),
© 2008 NRC Canada

394
2. (a) D.L. Boger, S.R. Duff, J.S. Panek, and M. Yasuda. J. Org.
Can. J. Chem. Vol. 86, 2008
Sharma, S. Sharma, and N. Rane. Biorg. Med. Chem. 12, 3135
(2004); (d) P. Sharma, A. Kumar, and M. Sharma. Eur. J.
Med. Chem. 41, 883 (2006); (e) P. Sharma, N. Rane, and P.
Pandey. Archiv. Pharm. 339, 572 (2006); (f) P. Sharma, A.
Kumar, N. Rane, and V. Guarram. Tetrahedron, 61, 4237
(2005).
Chem. 50, 5790 (1985); (b) D.L. Boger, J. Schumacher, M.D.
Mullican, M. Patel, and S.Panek, J. Org. Chem. 47, 2673
(1982); (c) H. Neunhoeffer and M. Bachmann. Chem. Ber.
108, 3877 (1975); (d) D.L. Boger, R.S. Coleman, J.S. Panek,
and D. Yohannes. J. Org. Chem. 49, 4405 (1984); (e) D.L.
Boger, R.S. Coleman, J.S. Panek, F.X. Huber, and J. Sauer. J.
Org. Chem. 50, 5377 (1985); (f) D.L. Boger and J.S. Panek.
Tetrahedron Lett. 24, 4511 (1983).
7. (a) M.J.S. Dewar, S. Olivella, and H.S. Rzepa. J. Am. Chem.
Soc. 100, 56 (1978); (b) M. Ortega, A. Oliva, J.M. Lluch, and
J.Bertran. Chem. Phys. Lett. 102, 317 (1983); (c) F.K. Brown
and K.N. Houk. Tetrahedron Lett. 25, 4609 (1985).
8. (a) L.R. Domingo, R.A. Jones, M.T. Picher, and J. Sepulveda
Arques. Tetrahedron, 51, 8739 (1995); (b) L.R. Domingo,
M.T. Picher, J. Andres, and V.S. Safont. J. Org. Chem. 62,
1775 (1997); (c) L.R. Domingo, M. Arno, and J. Andres. J.
Org. Chem. 64, 5867 (1999); (d) L.R. Domingo. Theor. Chem.
Acc. 104, 240 (2000); (e) L.R. Domingo. J. Org. Chem. 66,
3211 (2001).
9. (a) L.R. Domingo, M.T. Picher, J. Andres, V. Moliner, and
V.S. Safont. Tetrahedron, 52, 10693 (1996); (b) L.R. Domingo,
M.T. Picher, and R.J. Zaragoza. J. Org. Chem. 63, 9183
(1998); (c) L.R. Domingo, M.T. Picher, and M.J. Aurell. J.
Phys. Chem. A, 103, 11425 (1999); (d) L.R. Domingo, M.
Oliva, and J. Andres. J. Org. Chem.66, 6151 (2001); (e) L.R.
Domingo, J. Andres, and C.N. Alves. Eur. J. Org. Chem. 2557
(2002).
3. (a) D.L. Boger, D.R. Soenen, C.W. Boyce, M.P. Hedrick and
Q. Jin. J. Org. Chem. 65, 2479 (2000); (b) D.L. Boger and
S.E. Wolkenberg. J. Org. Chem. 65, 9120 (2000); (c) D.L.
Boger and J. Hong. J. Am. Chem. Soc. 123, 8515 (2001);
(d) A. Hamasaki, J.M. Zimpleman, I. Hwang, and D.L. Boger.
J. Am. Chem. Soc. 127, 10767 (2005).
4. (a) K. Fukui. Acc. Chem. Res. 4, 57 (1971); (b) R.B. Wood-
ward and R. Hoffmann. J. Am. Chem. Soc. 87, 4388 (1965);
see also the review: R.B. Woodward and R. Hoffmann. The
conservation of orbital symmetry. Verlag Chemie, Weinheim,
Germany. (1970); (c) G. Klopmann. J. Am. Chem. Soc. 90,
223 (1968); (d) L. Salem, J. Am. Chem. Soc. 90, 543, 553
(1968).
5. (a) G. Ujaque, J.E. Norton, and K.N. Houk. J. Org. Chem. 67,
7179 (2002); (b) L.R. Domingo, M.J. Aurell, M. Arno, and
J.A. Saez, J. Org. Chem. 72, 4220 (2007); (c) A. Zhong, C.
Rong, and S. Liu. J. Phys. Chem. A, 111, 3132 (2007);
(d) D.H. Ess and K.N. Houk. J. Am. Chem. Soc. 129, 10646
(2007); (d) P.R.Campodonico, P. Fuentealba, E.A.Castro, J.G.
Santos, and R. Contreras. J. Org. Chem. 70, 1754 (2005).
6. (a) P. Sharma, A. Kumar, and P. Pandey. Phosphorous Sulfur
Silicon Relat. Elem. 178, 583 (2003); (b) P. Sharma, A.
Kumar, and A. Mandloi. Synth. Commun. 33, 3 (2003); (c) P.
10. R.G. Parr, L. Von Szentpaly, and S. Liu. J. Am. Chem. Soc.
121, 1922 (1999).
11. L.R. Domingo, M.J. Aurell, P. Perez, and R. Contreras. Tetra-
hedron, 58, 4417 (2002).
12. MOPAC2007-DS-DG1, ©2007, Stewart Computational Chem-
istry.
© 2008 NRC Canada