Experimental and theoretical FMO interaction studies of the Diels-Alder reaction of 5-acetyl-3-methythio-1,2,4-triazine with cyclic enamines

Article abstract of DOI:10.1016/j.tet.2005.06.043

Diels-Alder reaction of 5-acetyl-3-methylthio-1,2,4-triazine with five cyclic enamines has been reinvestigated in its preparative and theoretical aspects. Its regioselectivity has been developed practically, which is in agreement with theoretical consideration of the FMO interactions, including secondary orbital interactions in the transition state. Since the energetic demands are similar for all five pairs, it has been indicated that their reactivity differences can be explained by an influence of steric hindrance in the considered transition state. In result, the synthesis of the 3-acetyl-1-methylthiocycloalka[c]pyridines, as synthons for preparation of sempervirine and its analogues has been optimized. The subsequent side reaction has been detected as a serious problem, especially in the case of the six-membered enamine, which reacts with the acetyl group of the final product formed in the reaction mixture.

Full text of DOI:10.1016/j.tet.2005.06.043

Tetrahedron 61 (2005) 8148–8158
Experimental and theoretical FMO interaction studies of the
Diels–Alder reaction of 5-acetyl-3-methythio-1,2,4-triazine
with cyclic enamines
*
Teodozja Lipi n´ ska
*
Institute of Chemistry, University of Podlasie, 3 Maja 54, 08-110 Siedlce, Poland
Received 23 March 2005; revised 31 May 2005; accepted 16 June 2005
Abstract—Diels–Alder reaction of 5-acetyl-3-methylthio-1,2,4-triazine with five cyclic enamines has been reinvestigated in its preparative
and theoretical aspects. Its regioselectivity has been developed practically, which is in agreement with theoretical consideration of the FMO
interactions, including secondary orbital interactions in the transition state. Since the energetic demands are similar for all five pairs, it has
been indicated that their reactivity differences can be explained by an influence of steric hindrance in the considered transition state. In result,
the synthesis of the 3-acetyl-1-methylthiocycloalka[c]pyridines, as synthons for preparation of sempervirine and its analogues has been
optimized. The subsequent side reaction has been detected as a serious problem, especially in the case of the six-membered enamine, which
reacts with the acetyl group of the final product formed in the reaction mixture.
q 2005 Elsevier Ltd. All rights reserved.
1. Introduction
In recent years aza-Diels–Alder reactions have been
important methods for the synthesis and transformation of
2
heterocyclic compounds. Many chemists have developed
applications of 1,2,4-triazines as azadienes in inverse
3
4
electron demand cycloadditions [4C2], both inter- and
Scheme 1. Scheme showing the two main ways in which the 1,2,4-triazine
ring can undergo Diels–Alder reaction.
5
intramolecular, for the synthesis of many functionalized
polyheterocyclic systems. The 1,2,4-triazine ring is known
to undergo two different sequences consisting in Diels–
3b,4a,8a,10
of the pyrimidine ring, (see Scheme 1b). In many
cases, the two modes of the reaction can run with the
3b,4a
formation of a mixture of products.
6
Alder–retro-Diels–Alder reactions (DA–rDA). The
sequence that begins with cycloaddition of an electron-
rich dienophile across C3–C6 of a 1,2,4-triazine moiety is
more common. Then follows the loss of N (rDA) and
Many reactions with different combinations of substituted 1,
2
2
,4-triazine-azadiene with an electron-rich dienophile have
elimination of a small molecule HX (HNR or HOR), which
2
11
been described recently. However, their frontier molecu-
lar orbital (FMO) interactions, which can indicate the route
of reaction, its regioselectivity and relationship between
structure of reactants and the rate of reaction have not been
investigated.
gives the pyridine ring as shown in Scheme 1a. Reactions of
2a,7b,8
with
7
the 1,2,4-triazine ring itself and its derivatives
cyclic enamines lead to the synthesis of 3,4-annulated
pyridines, important ring systems in pharmacologically
8
c,9
interesting molecules. The ynamine dienophiles usually
interact in the second sequence: cycloaddition across
N2–C5 with subsequent loss of nitrile (rDA) and formation
In our earlier investigation of the Diels–Alder reaction
between 3-substituted 5-acyl-1,2,4-triazine and enamines,
we demonstrated the synthetic usefulness of the replace-
ment of –N]N– fragment of 1,2,4-triazine by –C]C– of
*
See Ref. 1.
1
2
enamines for the preparation of alkyl-heteroaryl ketones.
Keywords: Aza Diels–Alder reaction; 1,2,4-Triazines; Frontier molecular
orbitals interactions; Secondary FMOs interplay in transition state;
Then we applied reactions between pairs 1 and 2a–d for
obtaining 3-acetyl-1-(methylthio)cycloalka[c]pyridines
3a–d, the key intermediates in the total synthesis of
3
*
-(Acetyl)-cycloalka[c]pyridines.
Tel.: C48 25 6431013; fax: C48 25 6442045; e-mail: tlip@ap.siedlce.pl
0040–4020/$ - see front matter q 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tet.2005.06.043

T. Lipi n´ ska / Tetrahedron 61 (2005) 8148–8158
8149
Scheme 2. Synthesis of the 3-acetyl-1-methylthiocycloalkene[c]pyridines 3a–d, as synthons of the sempervirine type alkaloids, via cascade transformation of
-acetyl-3-methylthio-1,2,4-triazine 1 with cyclic enamines.
5
pentacyclic indolo[2,3-a]quinolizine alkaloids; semper-
for full, quantitative consideration and optimization of the
transition states of reactions. We believe, however, that in
1
3
16
virine 4b (RZH, nZ2), its analogues with different E
1
4
rings 4a, 4c,d (RZH, nZ1, 3, 4) and next, their methoxy
the case of our investigated process, the basic study of the
FMO interactions is fundamental together with consider-
1
5
17
derivatives 5a–d (RZOCH , nZ1–4) (see Scheme 2).
3
This paper presents the correlation of the experimental data
and theoretical study of frontier molecular orbital inter-
actions of 1 and 2a–d in the Diels–Alder reaction in the
aspect of optimization of the synthesis of the final products
ing the possibility of emergence of the secondary (non-
bonding) interaction in the course of transition state
formation and it gives very interesting and satisfying
results.
1
8
3
a–d.
2.1. Preparative results
2
. Results and discussion
2.1.1. Experimental regioselectivity of the Diels–Alder
reaction of 1 with cyclic enamines 2a–e and structure
elucidation of products 3a–e. In our experiments directed
to the synthesis of condensed pyridines 3a–d in the Diels–
Alder reaction between 5-acetyl-3-methylthio-1,2,4-triazine
1 and enamines 2a–d it was visible (TLC plates) that the
consumption of the starting material was in all cases
accompanied by the extrusion of nitrogen. This indicated
that the first cycloaddition stage ran simultaneously with the
retro-Diels–Alder reaction, causing emergence and dis-
appearance of steric hindrance. In result, the two regio-
Differences in the course of the Diels–Alder reaction
between azadiene 1 and five–eight-membered cyclic
enamines 2a–d as dienophiles were observed during
optimization of these processes and, in result, the yields of
products 3a–d varied. The present paper contains the proof
of the correlation of the experimental data and the study of
the frontier molecular orbital (FMO) interactions in this
reaction. Because of the fact that the products 3a–d are
formed only in the first type of reaction (Scheme 1a), the
yields should depend on regioselectivity and the rate of
three-process cascade: cycloaddition of the enamines 2a–d
across C3–C6 positions of 1 (as the termini of azadiene),
0
isomeric dihydropyridine systems 6a–d or 6 a–d (Scheme
3) can be formed as intermediates in this step, due to the two
possible interaction of cyclic enamines. The second step
consists of elimination of pyrrolidine and formation of the
final products 3a–d, in which the structures are independent
of the regiochemistry of the dihydro-intermediates. We
observed, however, that the two visible steps had different
rates for enamines 2a–d. With the increase of the enamine
ring size, the rate of the first stage decreased and the rate
of the second one increased. Since the reaction between 1
and the enamine with a small ring (2a) proceeded quickly
towards a dihydropyridine-intermediate, which transformed
slowly into 3a, this intermediate compound could be
isolated from the cooled reaction mixture by preparative
retro-Diels–Alder reaction (involves the loss of N ) and
2
cis-elimination (XHZpyrrolidine) towards rearomatization
of the heterocyclic ring. These processes can be considered
as depending on the frontier molecular orbital (FMO)
interactions of reactants with the assumption that a
concerted transition state exists in the first stage of a
cascade reaction, and that emergence and disappearance of
steric hindrance in the course of the first and further
rearrangements play the most important role.
Recently, density functional theory (DFT) has been applied
Scheme 3. Cascade reaction between 1 and 2a–e.

8150
T. Lipi n´ ska / Tetrahedron 61 (2005) 8148–8158
thin-layer chromatography and its structure determined. In
the H NMR spectrum of this compound the signal of the
dihydropyridine proton appears at 6.68 ppm as a doublet
spectra of 3a–d, two three-proton singlets were presented
for H CO and H CS and two two-proton triplets or
1
3
3
multiplets for methylene groups bonded with pyridine
ring. Other methylene groups gave multiplets in the high-
field region. In this range the H NMR spectrum of 3e was
(
shown on the Scheme 3), but not 6 a. We observed (by TLC
JZ6.0 Hz), which indicates structure 6a (one enantiomer is
0
and H NMR spectra) the slow rearrangement of 6a into 3a
1
1
different and corresponded to its structure. The resonance
for C]O and HC3 of 3a–e was observed in the C NMR
1
3
in solvents during isolation and measurement at room
temperature. During GC/MS analysis, two substances were
also observed on the chromatogram with t Z11.9 and t Z
spectra at regions 200.0–201.0 and 113.0–118.0 ppm,
accordingly. The peak for H CCO appeared at 25.8–
R
R
3
1
3.8 in ratio 12:88, which gave mass spectra with molecular
26.2 ppm and for H CS at 12.4–13.8 ppm. The molecular
3
peaks for 3a (m/zZ207) and 6a (m/zZ278).
ion peaks were noticeable in the electron-impact mass
spectra of all products 3a–e. The base peaks (100%) were
%
assigned to the extrusion of SH (M-33) from the molecules
We performed the reaction of 1 with 1,2,5,6-tetrahydro-N-
methyl-4-(1-pyrrolidino)pyridine 2e as dienophile (see
Scheme 3) for the purpose of an additional regioselectivity
investigation, because the structure of final product of
cascade reaction in this case is influenced by regioselec-
tivity of the first stage. From the X-ray structure of the
product 3e (Fig. 1) as 3-acetyl-6-methyl-1-methylthio-
%
3a–c, CH (M-15) from the 3d and of NC H (M-44) from
%
3
2 6
the molecule 3e. The main fragmentation paths of the
molecules 3a–e could be defined thanks to investigation of
the MS spectra of these five molecules (see Scheme 4). In
the mass spectrum of 3a a three-peak sequence existed at
m/zZ118, 117 (M-90), 116, which indicated the existence
of the naked ring skeleton of this molecule. Interestingly
enough, in the spectra of 3b–e, these peaks were also present
(partial fragmentation of the aliphatic rings). In effect, the
mass spectra for all products 3a–e were identical in the
region below m/zZ116.
5
,6,7,8-tetrahydro-2,6-naphthyridine (not 2,7-naphthyridine
derivative, the second possible regioisomer), we deduced
regioselectivity of the dihydrointermediate 6e. This is in
agreement with our earlier investigations of 5,6,7,8-
tetrahydro-2,6-naphthyridine, which is substituted in the
1
same way as 3e, where in the H NMR spectrum the NOE-
effect in the resonance of HC4 (d 7.44) and H C5 (d 3.54)
was observed.
2
Scheme 4.
With the above mentioned results in hand, we started
investigating the conditions and limitations of our method
for the transformation of 1 into 3a–e.
2
.1.2. Optimization of the reaction conditions. We tested
many solvents for the reactions of 5-acetyl-3-methylthio-1,
,4-triazine 1 with enamines 2a–e. Some conditions and
2
yields are shown in Table 1. It was clear that for the
reactions of 1 with enamines 2a–d the best conditions were
in boiling ethanol. Reactivity of these dienophiles towards
products 2a–d was decreased when the large ring in the
dienophile molecules was increased. For the reaction with
enamine 2e, that has different properties, the best conditions
were in boiling dioxane. The courses of reactions were
observed on TLC plates. Thus, it seemed that in the reaction
of 1 with an equimolar amount of enamine 2a in boiling
solvents, complete consumption of substrate occurred
within 15–20 min. When heating was stopped, the dihydro-
intermediate could be isolated and its structure was
established as 6a. For complete rearrangement to com-
pounds 3a prolonged heating was necessary over 4–5 h
(method A1) or the reaction mixture had to be left at room
temperature for a longer period (method A3). The process of
the pyrrolidine elimination was faster (only 15–30 min)
when heating was continued in the presence of catalytic
amount (10%) of acetic acid (method A2). There were
smaller difficulties in elimination of pyrrolidine in the case
Figure 1. Molecular structure 3e with thermal ellipsoids drawn at 50%
probability level.
The two experiments discussed above show that in the
reaction of 5-acetyl-3-methylthio-1,2,4-triazine 1 with all
0
cyclic enamines 2a–e, interactions occur between the C1 of
dienophiles 2a–e and the C3 of the 1,2,4-triazine ring-diene
0
1
, and the C2 of enamines with C6 of the azadiene, leading
to formation of 6a–e as intermediates in DA–rDA reaction
sequences only (Scheme 3). These results are in agreement
with the theoretical FMO interaction study, which is also
described in this paper.
The structure of all products 3a–e was determined by
spectroscopic methods. A sharp absorption at a region
K1
1
686–1701 cm
in the IR spectra was assigned to the
1
C]O bond. The H NMR spectra of the 3a–e showed only a
single peak at a low field: 7.44–7.65 ppm, which indicated
one aromatic proton HC3. In the aliphatic region of the

T. Lipi n´ ska / Tetrahedron 61 (2005) 8148–8158
8151
a,b,c
Table 1. Optimization of the synthesis of 3a–e in the experimental scale 1 mmol of 1 in conventional heating in method A
0
(substrate 1 concentration
3
.2 mol/dm ), and under microwave irradiation in method B
d
Entry
Enamine
Solvent
Molar ratio
of 2a–e to 1
Reaction time
Product
Yield
(%)
For consumption of 1
in refluxing solvent
For complete transformation of the
dihydrointermediate 6a–e into final product
a
method A1 :5 h
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2a
2a
2a
2a
2a
2a
2b
2b
2b
2b
2b
2b
2c
2c
2c
2d
2d
2d
2d
2e
2e
2e
Benzene
Benzene
Benzene
Ethanol
Ethanol
Ethanol
Benzene
Benzene
Toluene
Ethanol
Ethanol
Ethanol
1
1
1
1
1
1
0.95
0.95
0.95
0.95
0.95
0.95
1.1
1.1
1.1
2
20 min
20 min
20 min
15 min
15 min
15 min
4 h
4 h
2 h
1.5 h
1.5 h
1.5 h
3 h
3a
3a
3a
3a
3a
3a
3b
3b
3b
3n
3b
3b
3c
3c
3c
3d
3d
3d
3d
3e
3e
3e
50
65
52
65
75
57
65
55
60
75
70
55
28
54
40
25
30
23
45
25
35
55
b
method A2 :20 min
c
method A3 :15 h
method A1:4 h
method A2:20 min
method A3:25 h
method A1:5 h
method A2:10 min
—
0
1
2
3
4
5
6
7
8
9
0
1
2
method A1:1.5 h
method A2:10 min
method A3:15 h
Toluene
Ethanol
—
—
—
—
—
—
—
—
—
—
2.5 h
d
method B ; 2.5 min
Chlorobenzene
Toluene
Ethanol
Ethanol
Chlorobenzene
Ethanol
5 h
15 h
2
e
2C1C1
3
1.5
1.5
1.5
25 h
d
method B ; 2.5 min
2 h
7 h
5 h
Toluene
Dioxane
a
b
c
d
e
Method A1: continuation of reflux in the same solvent.
Method A2: refluxing in the presence of acetic acid.
Method A3: left at room temperature.
Method B: fast microwave heating of the high concentration reaction mixture in chlorobenzene at temperature 110 8C.
Reaction without heating (room temperature 22 8C).
of 1 and 2b, and these problems were absent in cases of 1
and 2c–e, where new rings were larger. Products 3c–e were
formed simultaneously with disappearance of substrate 1 in
reaction with the appropriate enamine. Reaction between 1
and 1-pyrrolidine-1-cyclooctene 2d was very slow and gave
a low yield using conventional heating. It turned out that
the cascade process in this case was efficient when a high-
concentration reaction mixture in chlorobenzene (with
excess of enamine 2d) was heated by microwave irradiation
elimination, were faster for larger rings. All reactions did
not proceed cleanly. Different side products were noticed in
the reaction mixture (on TLC plates), especially when the
regime of process; the molar ratio of reactants and their
concentration, was not respected. For enamines 2a–c, the
concentration 1 in solvent had to be no higher than 0.2 mol/
3
dm . It is noteworthy than the high reactivity of enamines
2a,b (and to a small degree 2c) towards products 3a–b was
observed also as a tendency towards the formation of side
products in the cases when excess of the enamines was used,
and/or concentration of both reactants in solvent increased.
One kind of the side-product possessing the lowest polarity
(
1.5 min at 110 8C; method B). It means that in this case, the
first step of reaction, cycloadduct [2C4] formation, had to
run via the densely packed and hindered stage. In general,
the speed of the first step diminished together with
increasing the ring size of the enamines 2a–e ring, which
caused larger steric hindrances. However, the retro-Diels–
(the highest R ) was isolated from the reaction mixtures.
f
After determination of the structure of the representative
compound 7b (see Scheme 5), it became clear that they were
formed in reaction between products 3a–c and excess
Alder reactions, with N extrusion, and then pyrrolidine
2
Scheme 5. Proposed mechanism of the consecutive reaction between products 3a–c and appropriate enamine.

8152
T. Lipi n´ ska / Tetrahedron 61 (2005) 8148–8158
appropriate enamine 2a–c, which existed in the reaction
mixture. In Table 1 the optimal ratios of 2a–e to 1 for
minimization of this subsequent reaction are shown.
experiments, where pure 3a–c was reacted with appropriate
enamine 2a–c in refluxing ethanol (C in the experimental) or
toluene. We have noticed that the equilibrium was
established, when only trace concentration of compounds
7a–c existed. This balance remained in the presence of light,
but it moved towards the compounds 7a–c after pyrrolidine
was added. We have also detected in the reaction mixture
other isomeric products in trace quantities. They will be
investigated further. We aim at confirmation of existence of
the oxetanes B and at the general application of the reaction
of the heteroaromatic acetyl derivatives with cyclic
enamines.
The best reaction conditions were established in entry 5 for
3a, in entry 10 for 3a, in entry 14 for 3c, in entry 19 (under
microwaves) for 3d and in entry 22 for 3e.
It is interesting that the strongest tendency to consecutive
reaction was observed for 1 and 2b, while it was not
observed for 1 and 2d. The side product 7b was formed
during synthesis of 3b in all investigated methods and
solvents. The best yield of 3b (only a trace of 7b) was
obtained when 1 reacted with a deficiency of enamine 2b
and the concentration of reactants in solvents was no larger
2.2. Study of the frontier molecular orbitals (FMOs)
interactions
3
than 0.2 mol/dm . We noticed that in extreme conditions
(
B in experimental), when 1 reacted with double the amount
We studied the frontier molecular orbitals (FMOs) of the
5-acetyl-3-methylsulfanyl-1,2,4-triazine 1 as azadiene and
cyclic enamines 2a–e as dienophiles and considered their
interactions in a concerted Diels–Alder reaction as a first
step of a cascade rearrangement to the condensed pyridines
of 2b without solvent in room temperature, an exothermic
reaction was observed and compound 7b was isolated as a
major product (30%), together with only 20% of 3b and
many other side products more polar than these two.
3
observed regioselectivity and reactivity differences.
a–e. The FMOs model seemed capable of explaining the
1
7
The mechanism of formation of 7a–c in reaction of 3a–b
with the appropriate enamine is under investigation
Therefore, attempts were made to correlate the energies
and coefficients of the FMOs for these reactants using the
(
Scheme 5). In the first stage, oxetane formation is a
1
9
rational way for oxygen migration from the acetyl group to
cycloalkane ring, in anhydrous reaction medium, and it can
be formed in two ways. The first (route a) is preferred in
polar solvents and starts with the enamine carbanion
addition to acetyl group, as in the aldol type condensation.
The subsequent step is the intramolecular nucleophilic
attack of the oxygen on the imine carbon in the intermediate
A, which leads to oxetane B. Its formation in the second
possible way (route b) involves [2C2] photocyclization,
like in the Paterno-B u¨ chi reaction, and is preferred in non-
polar solvents in presence of light. However, both
mechanisms can take place as concurrently. The stereo-
chemistry of oxetane B (one enantiomer is shown on the
Scheme 5) can be determined in route a (intramolecular
nucleophilic addition), as well as in route b (concerted
cycloaddition). The second stage, oxetane rearrangement,
is, in summary, the cis-1,4-elimination of pyrrolidine. This
process can run as synchronous in six-membered transition
state, where three changes are simultaneous: formation
of methylene group, establishment of carbonyl group in
cycloalkane from the oxetane oxygen and elimination of
pyrrolidine via the intramolecular (or intermolecular)
accepting proton by nitrogen from the methyl group. It
seems that in the proposed mechanism, in the case of the
six-membered enamine 2b, steric hindrance does not exist,
and this reaction can run more easily than for 3a,c with
appropriate enamine 2a,c. An alternative mechanism via
cycloadduct C, which gives product 7b after dehydration
and enamine hydrolysis, is less probable because the
presence of an acidic reagent is necessary in this case.
However, intermediate oxetane B and hydroxyenamine C
have not been isolated so far.
semi empirical MOPAC-AM1 methodology. The graphs
of calculated energy of the HOMOs and LUMOs of 1
(exactly) and 2a–e (approximately) are shown in Figure 2.
Their p coefficient pictures and values also are shown. The
z
calculation suggests that the most efficient interaction,
leading to product formation, occurs between the LUMO of
the azadiene 1 (LUMO
2a–e (HOMO2a–e), not HOMO
means, it must be an inverse electron demand cycloaddition.
From the picture of the p coefficients of LUMO it is visible
) and the HOMOs of dienophiles
1
and LUMO2a–e. This
1
z
1
that either C3–C6 or N2–C5 positions can be the termini of
diene in molecule 1, which is in agreement with the two
known reactivity sequences of 1,2,4-triazines in Diels–
Alder reaction.
Figure 2. Calculated energy and the pictures (LCAO) of the FMOs (with p
coefficients and their values) of 1 and 2a–e, which interact in inverse
electron demand [4C2] cycloaddition.
z
If one takes into consideration the fact that the larger
coefficients of the FMOs of both reactants and the smaller
coefficients of both reactants combine, an asynchronous
transition state should be formed to predict the regio-
selectivity of cycloadduct. It turns out that there must be two
possibilities interactions in these two directions. They are
shown on the Figure 3, together with the calculated values
We deduced the mechanism discussed above from the fact
of isolation of 7b and its structure. In order to further
research the method of formation of the products 7a–c,
consisting in addition of enamine 2a–c to acetyl group in the
final products 3a–c (Scheme 5), we have recently carried out

T. Lipi n´ ska / Tetrahedron 61 (2005) 8148–8158
8153
Figure 3. The study of the four possibilities of interaction of the FMOs in Diels–Alder reaction between 1 and 2a–e: (a) two interaction ways of the enamines
with the positions C3–C6 of azadiene 1; (b) two interaction ways of the enamines with the positions N2–C5 of azadiene; (p-SOI, positive secondary orbitals
interaction; n-SOI, negative secondary orbitals interaction) shows the tendency to form mainly by the (a1) way of the sensitive 8a–e cycloadducts, which can
limit the cascade rearrangement.
of p coefficients of the LUMO and with the approximately
z
values of p coefficients of the HOMOs2a–e. The full values
z
of the latter are given in Table 2. The energy gaps between
the reactive frontier molecular orbitals are similar for all
effect on the course of reaction (n-SOI: negative SOI). In the
(b) situations the bonding interaction are as follows: the first
1
0
between C5 of azadiene 1 and C2 of dienophiles 2a–e
(bonding (3)), with larger p coefficients and the second
z
0
pairs 1 and 2a–e (see Table 2, DEZE(LUMO )K
between N2 of 1 and C1 of 2a–e (bonding (4), with smaller
1
E(HOMO2a–e)) and they give no reason for different course
of these reactions. Therefore, in search of a solution, we
considered precisely all four possible ways of formation of
the hypothetical transition state (see Fig. 3).
pz coefficients). These bonding interactions are stronger than
in case of the (a) combination because p (N2) and p (C5)
coefficients are larger and their overlap with p coefficients
z
z
z
of dienophiles 2a–e are bigger, than with p (C3) and p (C6)
z
z
coefficients in the (a) case. Aside from these, in case of the
(b1) exo-regiochemistry, there is only a weak bonding
interaction (p-SOI) between N1 of azadiene 1 and
pyrrolidine nitrogen. In the case of (b2) endo-combination
On Figure 3 it is visible that, when planar cyclic molecule 1
interacts with cyclic enamines 2a–e to form (a1), (a2), (b1),
(b2) possible transition states towards cycloaddition [4C2],
we must take into consideration the possibility of bonding
the reactants in the [4C2] cycloaddition, the p coefficients
z
interactions between p coefficients (1,2,3,4), which appear
z
of pyrrolidine nitrogen must interact in transition state with
two p coefficients: C3 and sulfur atom. The first of them has
a bonding nature (p-SOI), but the second one is anti-bonding
(n-SOI). In result, the reactants are, in both (b) cases, at a
disadvantage in the transition states. In Table 2 it is shown
at the ends of the azadiene and the dienophile molecules. In
addition, other than bonding, molecular interplay can exist.
Thus, it can be either beneficial and disadvantageous for
transition state secondary orbital interactions (SOI). In the
z
(
azadiene 1 and C2 of enamine 2a–e (bonding (1), with
larger p coefficients) and between C3 of 1 and C1 of
z
a) cases the main interactions appear between C6 of
0
that the differences between p coefficients for bonding
z
interactions in the (a) and (b) directions (Dp ) are similar
z
and do not play a decisive role in the course of transition
state formation and regioselectivity determination of the
cascade reaction.
0
enamine (bonding (2), with smaller p coefficients). The first
z
bonds are formed faster and are shorter than the second in
the (a1) and (a2) transition states. Thus, there are concerted
but asynchronous transition states. However, when these
two ways of interaction are considered, it is visible that in
the case of (a1), exo-regiochemistry of the Diels–Alder
To sum up the above considerations, we have come to
conclusion that the best situation for transition state
formation is in the case (a1). This exo-combination of the
azadiene moiety: C3–N4–C5–C6 and pyrrolidine nitrogen
as dienophile substituent results from the secondary orbitals
interaction, which can exist on the outside of the [4C2]
electronic system. This is the optimal starting point for
cascade transformation of compounds 1 in reaction with
cyclic enamines 2a–e. We can consider an increase in the
reactants, additional strong p -bonding interaction between
z
N2 of azadiene and pyrrolidine nitrogen of enamine may
occur. This secondary orbital interaction (SOI) is beneficial
for the (a1) transition state stabilization (p-SOI: positive
SOI). In the (a2) case (endo-regiochemistry), a small SOI
between the pyrrolidine nitrogen and N4 of 1 has an adverse

8154
T. Lipi n´ ska / Tetrahedron 61 (2005) 8148–8158
overlap of the orbitals forming the bonds as progress along
the reaction path. It means that the Diels–Alder cyclo-
adducts can have the 8a–e structure. They can be considered
as derivatives with the highest energy, which is necessary to
overcome the activation barrier. The strained transition
structures 8a–e must react quickly (exothermic process) to
give intermediates 6a–e after nitrogen extrusion (rDA). Our
considerations can be expanded in the situation when the
starting formation of the first transition state and the second
one run simultaneously. Then, after pyrrolidine elimination,
products 2a–e are formed. Steric hindrance increases in
course of the formation of the first transition state and in the
structure of cycloadducts 8a–e for enamines with larger
rings. This is the reason why the reactivity of hepta- and
octa-membered enamines is smaller than penta- and hexa-
membered, whereas the energetic demands are similar in all
these cases. The structure of the sensitive Diels–Alder
cycloadduct theoretically considered as 8a–e is compatible
with the experimentally established structure of the inter-
1
mediate 6a ( H NMR) and the final product 3e (X-ray).
3. Conclusion
In conclusion, this work contains correlations of the
experimental data towards optimization and the theoretical
calculations of the Diels–Alder reactions between 5-acetyl-
3
-methylthio-1,2,4-triazine 1 and cyclic enamines 2a–e.
We have demonstrated that the observed regioselectivity is
in agreement with the theoretical considerations of the
frontier molecular orbital interaction study, including
secondary interplay in the course of transition state
formation. Different reactivity of pairs of reactants, which
have similar energetic demands, can also be explained as the
influence of steric crowding of the considered transition
state. In result, we have developed the synthetic strategy for
the preparation of varied cycloalka[c]pyridines 3a–e. The
synthesized compounds contain the acetyl group in the C3
position, which opens access to further transformations,
such as the Fischer indole synthesis or aldol type
condensation.
4. Experimental
4
.1. General
-Acetyl-3-methylthio-1,2,4-triazine 1
from 3-methylthio1,2,4-triazine, via a two-step procedure
1
3,20
5
was prepared
2
1
1
3
described previously. All reactions were performed
under calcium chloride tube, in anhydrous solvents, which
2
2
were dried in standard procedures prior to use. Enamines
a–e were synthesized from commercial ketones and
2
2
3
pyrrolidine (Aldrich) by Kuehne method. Microwave
reactor Synthewave 402 (Prolabo, 300 W, focused micro-
waves, open, rotating system of reaction vessel) with
software (feedback temperature monitoring) was used.
The course of reactions was monitored by thin-layer
chromatography (TLC), which was carried out on
0
.25 mm Merck silica gel plates (60F254). Column
chromatography was performed on Merck silica gel 60
230–400 mesh). Melting points were determined on
Bo e¨ tius microscopic plate and are uncorrected. All new
(

T. Lipi n´ ska / Tetrahedron 61 (2005) 8148–8158
8155
1
compounds were determined to be O95% pure by H NMR.
IR spectra (KBr pellets) were recorded on FT-IR Magna 760
(
(KBr): 3072, 2956, 2928, 2835, 1686, 1574, 1556, 1426,
1281, 1251, 1170, 923, 905, 581 cm ; EI-MS m/z (%): 207
K1
C
(M , 50), 192 (10), 174 (100), 164 (8), 156 (9), 149 (12),
Nicolet) apparatus. Mass spectra and high-resolution
measurements were obtained with an AMD 604 (Intectra,
GmbH, Germany) spectrometer. GC/MS experiments were
recorded on GC gas chromatograph-MS-QP550 mass
detector (Shimadzu) with Zebron ZB-5, 30 M!0.25 mm
ID!0.10 mM column. H and C NMR spectra were
recorded with Varian Gemini (200 MHz) and Mercury
132 (8), 118 (5), 117 (4), 116 (5), 105 (4) 91 (3), 77 (7), 65
(7), 51 (8), 43 (10); HRMS (EI, M ) calcd for C H NOS
C
1
1 13
207.0718, found 207.0717. Anal. Calcd for C H NOS: C,
1
1 13
63.76; H, 6.28; N, 6.76; Found: C, 64.04; H, 6.40; N, 6.75.
1
13
4.2.2. 1-(1-Methylthio-5,6,7,8-tetrahydro-isoquinolin-3-
yl)-ethanone (3b). Colorless crystals with mp 68–69 8C
(from dichloromethane/hexane), described in our previous
4
obtained with Perkin-Elmer 2400-CHN analyzer. The AM1
00BB (400 MHz) spectrometers. Elemental analyses were
1
9a
19b
24
method
ware was employed for semi empirical calculations.
from the MOPAC,
CAChe 5.0 Fujitsu soft-
paper. mp 62–63 8C (after sublimation). The new data;
1
9c
1
R Z0.45 (dichloromethane); H NMR (400 MHz, CDCl ):
f
3
d 1.74–1.82 (2H, m), 1.84–1.92 (2H, m), 2.60 (2H, t, JZ
6.6 Hz), 2.61 (3H, s),2.70 (3H, s), 2.76 (2H, t, JZ6.1 Hz),
7.48 (1H, s); C NMR (100 MHz, CDCl ): d 12.99 (SCH ),
3 3
4.2. General procedures for Diels–Alder reaction
between 1 and 2a–e
1
3
2
1.87(CH ), 22.87 (CH ), 25.32 (CH ), 25.83 (H CCO),
2
2
2
3
Method A. To a solution of 1 (1 mmol) in ethanol or other
solvent listed in the Table 1 (5 mL) was added an enamine
29.27 (CH ), 117.96 (CH), 134.40 (C), 145.81 (C), 149.75
2
(C), 158.28 (C), 200.56 (CO); IR (KBr): 3064, 2933, 2920,
2866, 1691, 1579, 1549, 1415, 1394, 1357, 1287, 1240,
2a (1 mmol) or 2b (0.95 mmol) or 2c (1.1 mmol) or 2d
2 mmol) or 2e (1.5 mmol). The reaction mixture was
K1
1174, 967, 951, 843, 819, 589 cm ; EI-MS m/z (%): 221
(
refluxed under calcium chloride tube for the moment
substrate 1 disappearing (monitored by TLC).
C
(M , 51), 206 (20), 188 (100), 170 (13), 160 (5), 146 (7),
130 (10), 117 (6), 103 (6), 91 (7), 77 (13), 65 (5), 51 (13), 43
(25).
In the case of reaction between 1 and 2a, and 1 and 2b, an
intermediate existed in the reaction mixture. Their complete
transformation into products 3a or 3b was carried out with
three ways: A1: continuation of reflux in the same solvent;
A2: acetic acid (0.10 mmol) was added to the reaction
mixture, which was refluxed for 20 min in the case of 3a and
4.2.3. 1-(1-Methylthio-6,7,8,9-tetrahydro-5H-cyclohepta-
[c]pyridin-3-yl)-ethanone (3c). Colorless crystals with mp
56–57 8C, (from dichloromethane/hexane); R Z049
f
1
(dichloromethane); H NMR (400 MHz, CDCl ): d 1.59–
1.67 (4H, m), 1.83–1.92 (2H, m), 2.58 (3H, s), 2.69 (3H, s),
3
1
3
10 min in the case of 3b; A3: the reaction mixture was
stirred at room temperature.
2.77–2.83 (2H, m), 2.87–2.92 (2H, m), 7.50 (1H, s);
C
NMR (100 MHz, CDCl ): d 13.85 (SCH ), 25.80 (H CCO),
3
3
3
26.06 (CH ), 26.88 (CH ), 30.00 (CH ), 32.28 (CH ), 35.72
(CH ), 118.17 (CH), 140.08 (C), 150.62 (C), 152.26 (C),
156.76 (C), 200.34 (CO); IR (KBr): 3066, 2945, 2921, 2849,
1701, 1574, 1545, 1445, 1417, 1393, 1350, 1350, 1291,
2 2 2 2
Products 3c–e were formed simultaneously with slow
disappearance of substrate 1 in reaction with an excess of
appropriate enamine 2c–e in refluxing solvent.
2
K1
261, 1194, 1149, 956, 920, 881, 843, 591 cm ; EI-MS
1
C
In all reactions the solvent was removed under reduced
pressure and product was isolated by silica gel column
chromatography (eluent: 1:1 hexane/dichloromethane for
3a–d and dichloromethane to 10:1 dichloromethane/acetone
for 3e).
m/z (%): 235 (70, M ), 220 (65), 206 (25), 202 (100), 192
(20), 174 (65), 160 (18), 144 (9), 130 (7), 118 (9), 117 (8),
116 (8), 105 (5), 91 (10), 77 (10), 65 (6), 51 (11), 43 (12).
Anal. Calcd for C H NOS: C, 66.34; H, 7.28; N,5.95;
1
3 17
Found: C, 66.20; H, 7.35; N,5.85.
Method B. To a solution of 1 (1 mmol) in chlorobenzene
(
4.2.4. 1-(1-Methylthio-5,6,7,8,9,10-hexahydro-cycloocta-
[c]pyridin-3-yl)-ethanone (3d). Colorless crystals with mp
0.5 mL) placed in a Pyrex cylindrical vessel was added
enamine 2c (1.1 mmol) or 2d (2 mmol). The reaction
mixture was irradiated in the Synthewave 402 microwave
reactor. The temperature setpoint was programmed at
101–102 8C, (from dichloromethane/hexane); R Z0.52
f
1
(dichloromethane); H NMR (400 MHz, CDCl ): d 1.24–
1.42 (4H, m), 1.62–1.84 (4H, m), 2.62 (3H, s), 2.71 (3H, s),
3
1
3
1
10 8C. Irradiation was stopped after 3.0 min from the
2.75–2.81 (2H, m), 2.88–2.94 (2H, m), 7.54 (1H, s);
C
NMR 100 MHz, CDCl : d 13.51 (SCH ), 25.82 (H CCO),
moment the temperature began to rapidly increase. The
reaction mixture was left to chill to 90 8C then was cooled to
room temperature. Product 3c or 3d was isolated by silica
gel column chromatography (1:1 dichloromethane/hexane).
3
3
3
25.86 (CH ), 26.01 (CH ), 27.03 (CH ), 28.10 (CH ), 31.46
2
2
2
2
(CH ), 32.37 (CH ), 118.25 (CH), 137.72 (C), 150.36 (C),
2
2
150.75 (C), 157.40 (C), 200.52 (CO); IR (KBr): 3064, 2955,
918, 2845, 1694, 1573, 1547, 1386, 1350, 1295, 1264,
2
K1
4
.2.1. 1-(1-Methylthio-6,7-dihydro-5H-[2]pyrindin-3-yl)-
ethanone (3a). Colorless crystals with mp 68.5–69.5 8C
from dichloromethane/hexane); R Z0.43 (dichloro-
1195, 1159, 1082, 1014, 899, 839, 598 cm ; EI-MS m/z
(%): 249 (60, MC), 234 (100), 220 (25), 216 (32), 206 (31),
202 (28), 188 (25), 174 (12), 162 (5), 144 (4), 130 (9), 118
(9), 117 (8), 116 (9), 105 (6), 91 (10), 77 (12), 65 (9), 51
(
f
1
methane); H NMR (400 MHz, CDCl ): d 2.15 (2H, quintet,
JZ7.6 Hz), 2.64 (3H, s), 2.70 (3H, s), 2.82 (2H, t, JZ
3
(12), 43 (20); HRMS (EI, MC) calcd for C14H19NOS
249.1184, found 249.1192.
1
3
7
.6 Hz), 2.94 (2H, t, JZ7.6 Hz), 7.64 (1H, s); C NMR
(
(
(
100 MHz, CDCl ): d 12.41 (SCH ), 24.00 (CH ), 26.02
H CCO), 30.23 (CH ), 32.64 (CH ), 113.74 (CH), 141.01
C), 151.73 (C), 153.38 (C), 154.16 (C), 200.09 (CO); IR
3
3
2
4.2.5. 1-(1-(Methylthio)-5,6,7,8-tetrahydro-2,6-naph-
thyridin-3-yl)-ethanone (3e). Yellow crystals with mp
3
2
2

8156
T. Lipi n´ ska / Tetrahedron 61 (2005) 8148–8158
1
02.5–103 8C, (from dichloromethane/diethyl ether,
chromatography using dichloromethane/hexane 1:2 to 1:1
1
anhydrous); R Z0.31 (3:1 dichloromethane/acetone); H
NMR (400 MHz, CDCl ): d 2.45 (3H, s), 2.62 (3H, s), 2.70
as eluent. Pure compound 7b was obtained (60 mg, 20%)
and then pure 3b (44 mg, 20%). The purity of products were
analyzed by TLC (dichloromethane; R for 1, 3b, 7b was
f
3
(
(
(
(
3H, s), 2.74 (4H, s), 3.54 (2H, s), 7.44 (1H, s); NMR
100 MHz, CDCl ): d 12.85 (SCH ), 25.79 (H CCO), 25.88
NCH ), 45.71 (CH ), 51.89 (CH ), 57.07 (CH ), 115.21
3
f
1
measured as 0.11, 0.45 and 0.70), GC/MS and H NMR.
3
3
3
2
2
2
CH), 131.56 (C), 143.52 (C), 150.22 (C), 158.24 (C),
B. Reaction between 1 and 2b in molar ratio 1:2 without
solvent. To pure 1 (169 mg, 1 mmol) under argon was added
dropwise 2b (2 mmol) at room temperature during 5 min.
Vigorous nitrogen extrusion and temperature increase to
40 8C occurred. Complete consumption of substrate 1 was
observed (TLC monitoring using dichloromethane) after
stirring for 30 min. The reaction mixture was resolved by
column chromatography using dichloromethane/hexane 1:2
2
1
1
2
00.20 (CO); IR (KBr): 3066, 2966, 2920, 2837, 2774,
690, 1580, 1555, 1415, 1367, 1348, 1291, 1247, 1177,
K1
126, 1061, 951, 902, 858, 809, 590 cm ; EI-MS m/z (%):
C
36 (70, M ), 221 (54), 203 (33), 193 (37), 192 (100), 189
(
1
8), 178 (4), 174 (15), 160 (4), 150 (11), 136 (4), 118 (4),
17 (3), 104 (3), 91 (3), 77 (4), 65 (4), 51 (6), 43 (7); HRMS
C
(
EI, M ) calcd for C H N OS 236.0983, found 236.0980.
12 16 2
The X-ray structure is shown (Fig. 1) and its data are listed
below.
to 1:1 as eluent. Compound 7b (90 mg, 30%) with R Z0.70
f
(dichloromethane) and t Z15.8 min (base peak m/zZ301)
R
was first isolated and then an intermediate fraction (35 mg,
12%), which contained 7b and a trace of compound
4.3. Isolation and structure determination of inter-
mediate of 6a
isomeric to 7b with R
6.2 min (base peak m/zZ301) and next pure 3b (44 mg,
20%).
Z0.62 (dichloromethane) and t Z
f
R
1
The mixture of 5-acetyl-3-methylthio-1,2,4-triazine 1
(
42 mg, 0.25 mmol) and 1-(1-cyclopenen-1-yl)-pyrrolidine
a (0.25 mmol) in ethanol (2 mL) was refluxed for 15 min
and cooled to 0 8C. TLC monitoring (50:1 dichloromethane/
2
C. Reaction between 3b and 2b in ethanol. To the solution
of 3b (110 mg, 0.5 mmol) in ethanol (2.5 mL) was added
1-(1-cyclohexen-1-yl)-pyrrolidine 2b (0.5 mmol). The
reaction mixture was refluxed under nitrogen for 3 h.
It was observed (TLC monitoring, dichloromethane) that
only trace concentration of 7b was present in the reaction
mixture besides starting material 3b. Next, pyrrolidine
(0.5 mmol) was added to the reaction mixture and reflux
was continued for 3 h. After cooling, the solvent was
evaporated under reduced pressure. The residue was
resolved by column chromatography to afford: 7b (39 mg,
26%), then an intermediate fraction (28 mg, 10%), which
contained 7b and its isomer in the molar ratio 1:2.5 (by GC/
MS) and 3b (17 mg, 15%). The purity of products were
acetone) showed that the substrate 1 (R Z0.65) disappeared
f
but only a trace of product 3a (R Z0.75) existed in the
f
reaction mixture besides predominantly the most polar
compounds (R Z0.28). The latter was isolated by prepara-
f
tive thin-layer chromatography using 50/1 dichloro-
methane/acetone as mobile phase. A white substance
(
0
53 mg, w75%) was obtained, which was stored below
8C. In GC/MS measurements of this substance were
observed two compounds: 3a and 6a on the chromatogram
in ratio 12:88 with retention times t Z11.9 and t Z13.8,
R
R
respectively, which gave MS spectra: for 3a m/z (%): 207
(
C
M , 40), 192 (6), 174 (100), 164 (5), 156 (8), 149 (5), 132
1
(
5
8), 118 (5), 117 (4), 116 (5), 105 (4), 91 (8), 77 (8), 65 (7),
analysed by TLC (dichloromethane), GC/MS and H NMR.
C
1 (9), 43 (80); for 6a m/z (%): 278 (M , 9), 263 (57), 231
(
1
(
97), 209 (6), 203 (13), 188 (15), 174 (10), 162 (10), 148 (7),
36 (100), 118 (11), 96 (9), 91 (16), 70 (47), 55 (39), 43
85). The H NMR spectrum also showed the presence of
4.4.1. 2-[1-(1-Methylthio-5,6,7,8-tetrahydro-isoquinolin-
3-yl)-vinyl]-cyclohexanone (7b). Colorless crystals with
1
mp 52–53 8C, (from dichloromethane/hexane); R Z0.70
f
two compounds in the ratio 9:1, which was calculated from
the integration ratio of peaks: 6.68 ppm (1H, d, JZ6 Hz, for
(dichloromethane); GC/MS: 97/3 mixture of 7b/its isomer
was observed on the chromatogram with retention times
t Z15.8 (95%) and t Z16.2 (3% m/zZ301). MS spectrum
for 7b m/z (%): 301 (M , 67), 286 (95), 284 (45), 272 (78),
258 (60), 244 (45), 240 (61), 231 (77), 226 (40), 212 (10),
206 (13), 198 (11), 193 (28), 178 (100), 163 (12), 150 (12),
6
a HC4) and 7.64 ppm (1H, s, HC4 for 3a). This spectrum
R
R
C
also showed two singlets at 2.43 and 2.48 ppm for resonance
of two methyl groups of 6a. Other aliphatic protons (15H) of
this substance gave wide-ranging multiplets in 2.9–1.5 ppm
region.
144 (16), 130 (30), 117 (18), 103 (22), 91 (26), 77 (33), 53
(
(4H, m), 1.72–1.82 (2H, m), 1.83–1.91 (2H, m), 1.98–2.05
1
31), 41 (91); H NMR (400 MHz, CDCl ): d 1.52–1.68
3
4.4. Isolation and structure determination of side
product 7b
(4H, m), 2.61 (2H, t, JZ6.5 Hz), 2.62 (3H, s), 2.76 (2H, t,
1
3
JZ5.6 Hz), 3.85 (2H, s), 5.56 (1H, br s), 7.48 (1H, s);
C
NMR (100 MHz, CDCl ): d 13.12 (SCH ), 21.88 (CH ),
A. Reaction between 1 and 2b in molar ratio 1:2 in ethanol.
To a solution of 5-acetyl-3-methylthio-1,2,4-triazine 1
3
3
2
22.11 (CH ), 22.41 (CH ), 22.86 (CH ), 25.41 (CH ), 28.54
2
2
2
2
(
169 mg, 1 mmol) in ethanol (5 mL) was added 1-(1-
(CH ), 29.29 (CH ), 29.68 (CH ), 46.06 (CH), 118.47 (CH),
2 2 2
cyclohexen-1-yl)-pyrrolidine 2b (2 mmol). The reaction
mixture was refluxed for 5 h and cooled to room
temperature. TLC monitoring (50:1 dichloromethane/
125.50 (]CH ), 132.40 (C), 134.38 (C), 145.89 (C), 149.62
2
(C), 158.14 (C), 200.44 (CO); IR (KBr): 3068, 2929, 2853,
2830, 1692, 1576, 1548, 1426, 1402, 1390, 1328, 1283,
1224, 1141, 1033, 997, 843, 762, 613; EI-MS m/z (%): (301
acetone) showed that substrate 1 (R Z0.65) disappeared
f
C
but besides the product 3b (R Z0.77) the less polar
M , 100), 286 (72), 284 (45), 273 (32), 272 (57), 258 (48),
f
compound 7b (R Z0.90) existed in the reaction mixture.
After removing solvent under reduced pressure, careful
isolation of products was achieved by column
244 (30), 240 (46), 231 (60), 226 (28), 212 (72), 206 (12),
198 (7), 193 (24), 178 (70), 163 (9), 150 (7), 135 (7), 117
(10), 116 (7), 103 (11), 91 (8), 79 (6), 77 (35), 67 (29), 53
f

T. Lipi n´ ska / Tetrahedron 61 (2005) 8148–8158
8157
C
6), 41 (15); HRMS (EI, M ) calcd for C H NOS
(
G.; Huber, F. X.; Sauer, J. Tetrahedron Lett. 1983, 24,
1481–1484. (c) Mueller, K.; Sauer, J. Tetrahedron Lett. 1984,
1
8 23
3
01.15004, found 301.14846.
2
5, 2541–2544. (d) Barlow, M. G.; Haszeldine, R. N.;
Simpkin, D. J. Chem. Soc., Perkin Trans. 1982,
245–1249. (e) Benson, S. C.; Gross, J. L.; Snyder, J. K. J.
1
4
.5. Crystal data for 3e
1
Org. Chem. 1990, 55, 3257–3269. (f) Sauer, J. Bull. Soc. Chim.
Belg. 1992, 101, 521–540.
C H N O S: MZ262.41, crystal dimensions 0.60!0.
1
2 26 2 2
3
0!0.30 mm , monoclinic, space group P 21/c (no. 14),
4
˚
˚
˚
5. (a) Taylor, E. C.; Macor, J. E. Tetrahedron Lett. 1985, 26,
2415–2418. (b) Seitz, G.; George, L.; Dietrich, S. Tetrahedron
Lett. 1985, 26, 4355–4358. (c) Taylor, E. C.; Pont, J. L. J. Org.
Chem. 1987, 52, 4287–4292. (d) Taylor, E. C.; Macor, J. E.
J. Org. Chem. 1987, 52, 4287–4292. (e) Charushin, V. N.; van
Veldhuizen, B.; van der Plas, H. C.; Stam, C. H. Tetrahedron
aZ7.9070(3) A, bZ14.7600(7) A, cZ21.5930(8) A, bZ
3
˚
1
0
10.477(3)8, UZ2360.82(17) A , ZZ4, F(000)Z295, D Z
c
3 K1
.738 g m , TZ100(2) K, m(Mo Ka)Z4.19 mm , Nonius
Kappa-CCD diffractometer, qmaxZ24.718, 3065 unique
reflections, which were used in all calculations. The
structure was solved by direct methods using the
2
4
1
989, 45, 6499–6512. (f) Taylor, E. C.; Macor, J. E.; French,
SHELXS-97 program was refined by full matrix least-
2
25
L. G. J. Org. Chem. 1991, 56, 1807–1812. (g) Benson, S. C.;
Li, J.; Snyder, J. K. J. Org. Chem. 1992, 57, 5285–5287. (h) Li,
J.; Snyder, J. K. J. Org. Chem. 1993, 58, 516–519.
squares on F using the program SHELXL-97. H-atoms
were included in idealized positions and refined isotropi-
cally. Refinement converged at R1Z0.0553, wR2Z0.0908
for all data and 295 parameters (R1Z0.0418, wR2Z0.0849
for 2569 reflections with I O2s(I )). The goodness-of-fit on
6
7
8
. Rickborn, B. In The Retro Diels–Alder Reaction. Part II.
Dienephiles with One or More Heteroatom; Paguette, L. A.
et al., Eds.; Organic Reactions; Wiley, 1998; Vol. 53, Chapters
I.D. and IV.D. with Table and References cited therein.
. (a) Boger, D. L.; Panek, J. S. J. Org. Chem. 1981, 46,
2179–2182. (b) Boger, D. L.; Panek, J. M.; Meier, M. M.
J. Org. Chem. 1982, 47, 895–897. (c) Boger, D. L.; Patel, M.;
Mulican, M. D. Tetrahedron Lett. 1982, 4559–4563.
o
o
2
2
2
o
F was equal 1.068. A weighting scheme wZ[s (F C
2
K1
2
2
(
in the final stage of refinement. The residual electron
0.0418P) C3.1964P)] where PZ(F C2F )/3 was used
o c
K3
˚
densityZ0.20/K0.21 e A
.
excluding structure factors) for the structure reported in
Crystallographic data
(
this paper have been deposited with the Cambridge
Crystallographic Data Centre as supplementary publication
no. CCDC-264189. Copies of the data can be obtained free
of charge on application to CCDC, 12 Union Road,
Cambridge CB21EZ, UK (fax: C44 1223 336 033;
e-mail: deposit@ccdc.cam.ac.uk).
. (a) Neunhoeffer, H.; Metz, H.-J. Justus Liebiegs Ann. Chem.
1
983, 1476–1496. (b) Taylor, E. C.; McDaniel, K. F.; Warner,
J. C. Tetrahedron Lett. 1987, 28, 1977–1980. (c) Taylor, E. C.;
Warner, J. C. Heterocycles 1987, 26, 2673–2685. (d) Chenard,
B. L.; Ronau, R. T. J. Org. Chem. 1988, 53, 5175–5177.
(
e) Taylor, E. C.; Macor, J. E. J. Org. Chem. 1989, 54,
249–1256.
. (a) Ohba, M.; Izuta, R.; Shimizu, E. Tetrahedron Lett. 2000,
1, 10251–10255. (b) Hautefaye, P.; Cimeti e` re, B.; Pierr e´ , A.;
1
9
Acknowledgements
4
L e´ once, S.; Hickman, J.; Laine, W.; Bailly, C.; Lavielle, G.
Bioorg. Med. Chem. Lett. 2003, 13, 2731–2725.
The author is grateful to Stefan J. Czarnocki and Aleksandra
Pawełkiewicz, students of the University of Podlasie, for
technical assistance.
1
1
0. Neunhoeffer, H.; Lehman, B. Justus Liebigs Ann. Chem. 1977,
413–1420.
1. (a) Wan, Z.-K.; Snyder, J. K. Tetrahedron Lett. 1998, 39,
487–2490. (b) Sauer, J.; Heldman, D. K.; Pabst, G. R. Eur. J.
1
2
Org. Chem. 1999, 313–321. (c) Lipi n´ ska, T.; Branowska, D.;
Rykowski, A. Khim. Geterosikl. Soedin. 1999, 381–389.
References and notes
(
d) Stanforth, S. P.; Tarbit, B.; Watson, M. D. Tetrahedron
1
2
. Part 32 in a series ‘1,2,4-Triazines in Organic Synthesis’. For
Part 31, see: Branowska, D.; Rykowski, A. Tetrahedron Lett.,
in print.
Lett. 2002, 43, 6015–6017. (e) Branowska, D. Synthesis 2003,
2089–2094. (f) Kozhevnikov, V. N.; Kozhevnikov, D. N.;
Nikitina, T. V.; Rusinov, V. L.; Chupakhin, O. N.; Zabel, M.;
K o¨ nig, B. J. Org. Chem. 2003, 68, 2882–2888. (g) Lahue, B.
R.; Wan, Z.-K.; Snyder, J.K J. Org. Chem. 2003, 68,
4345–4354. (h) Branowska, D. Tetrahedron 2004, 60,
6021–6027. (i) Raw, S. A.; Taylor, R. J. K. Tetrahedron
Lett. 2004, 45, 8607–8610. (j) Raw, S. A.; Taylor, R. J. K.
J. Am. Chem. Soc. 2004, 126, 12260–12261.
. For rewiews, see: (a) Boger, D. L.; Weinreb, S. N. In
Wasserman, H. H., Ed.; Hetero Diels–Alder Methodology in
Organic Synthesis; Academic: New York, 1987; Vol. 47.
(
b) Kametani, T.; Hibino, S. In Katritzky, A. R., Ed.; Advances
in Heterocyclic Chemistry; Academic: New York, 1987; Vol.
2, Chapters 9 and 10, pp 245–333. (c) Boger, D. L. Bull.
4
Chim. Soc. Belg. 1990, 99, 599–615. (d) Neunhoeffer, H. In
Katritzky, A. R., Rees, C. W., Scriven, E. F. V., Eds.;
Comprehensive Heterocyclic Chemistry II; Pergamon:
Oxford, 1996; Vol. 6, Chapters 9 and 10, pp 507–574.
12. Rykowski, A.; Lipi n´ ska, T. Pol. J. Chem. 1997, 71, 83–90.
13. Rykowski, A.; Lipi n´ ska, A. Synth. Commun. 1996, 26,
4409–4414.
14. Lipi n´ ska, T. Tetrahedron Lett. 2002, 43, 9565–9657.
15. Lipi n´ ska, T. Tertahedron Lett. 2004, 45, 8331–8334.
16. (a) McCarric, M. A.; Wu, Y. D.; Houk, K. N. J. Am. Chem.
Soc. 1992, 114, 1499–1500. (b) Cislowski, J.; Sauer, J.;
Hetzenegger, J.; Karcher, T.; Hierstetter, T. J. Am. Chem. Soc.
1993, 115, 1353–1359.
(
e) Boger, D. L. J. Heterocycl. Chem. 1998, 35, 1001–1011.
3
. For preliminary studies, see: (a) Dittmar, W.; Sauer, J.; Steigel,
A. Tetrahedron Lett. 1969, 5171–5174. (b) Steigel, A.; Sauer,
J. Tetrahedron Lett. 1970, 3357–3360. (c) Neunhoeffer, H.;
Fr u¨ hauf, H.-W. Liebigs Ann. Chem. 1972, 758, 121–131.
. For example, see: (a) Reim, H.; Steigel, A.; Sauer, J.
Tetrahedron Lett. 1975, 2901–2904. (b) Balear, J.; Chrisam,
4
17. (a) Fleming, I. Frontiers Orbitals and Organic Chemical
Reactions; Wiley: London, 1976. (b) Rauk, A. Orbital

8158
T. Lipi n´ ska / Tetrahedron 61 (2005) 8148–8158
Interaction Theory in Organic Chemistry; Wiley: New York,
994. (c) Fleming, I. Pericyclic Reactions; Oxford University
Press: Oxford, New York, Tokyo, 1999.
20. Rykowski, A.; Olender, E.; Branowska, D.; van der Plas, H. C.
Org. Prep. Proced. Int. 2001, 33, 501–503.
1
21. (a) Neunhoeffer, H.; Henning, H.-W.; Fr u¨ hauf, M.; Mutterer,
M. Tetrahedron Lett. 1969, 3147–3150. (b) Paudler, W. W.;
Chen, T. K. J. Heterocycl. Chem. 1970, 7, 767–770.
22. Perrin, D. D.; Armarego, W. L. F. Purification of Laboratory
Chemicals; Pergamon: Oxford, 1992.
1
8. For role of secondary orbitals interaction, see: (a) Ginsburg, D.
Tetrahedron 1983, 39, 2095–2135. (b) Juric, B. S. J. Chem.
Soc., Perkin Trans. 2 1999, 373–378. (c) Garcia, J. I.; Mayoral,
J. A.; Salvatella, L. Acc. Chem. Res. 2000, 33, 658–664.
9. (a) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J.
P. J. Am. Chem. Soc. 1985, 107, 3902–3909. (b) Stewart, J. J.
P. J. Comput. Aided Mol. Des. 1990, 4, 1–105. (c) Stewart, J. J.
P. Fujitsu Limited, Tokyo; FQS Poland, MOPAC 2002 in
CACHe 5.0 software.
1
23. Kuehne, M. E. J. Am. Chem. Soc. 1959, 81, 5400–5404.
24. Sheldrick, G. M. Acta Crystallogr., Sect. A 1990, 46, 467–473.
25. Sheldrick, G. M. SHELXL 97: Program for Refinement of
Crystal Structures; University of G o¨ ttingen: G o¨ ttingen,
Germany, 1993.