On-Demand Selection of the Reaction Path from Imino Diels-Alder to Ene-Type Cyclization: Synthesis of Epiminopyrimido[4,5-b]azepines

Article abstract of DOI:10.1002/ejoc.201301318

Controlling the mode of reaction of a reactive intermediate such as an imine or iminium ion should enable the on-demand selection of the final products from the same starting materials. The successful execution of such a strategy will reduce the time required to prepare diverse scaffolds. The imines derived from 4-(allylamino)pyrimidine-5-carbaldehydes and anilines undergo Diels-Alder reactions to give pyrimido[4,5-h][1,6]naphthyridines in high yields. A complete switch from the intramolecular aza-Diels-Alder (IADA) path to an ene-type cyclization reaction was achieved by simply adjusting the reaction conditions (amount of acid catalyst, solvent, and temperature). This newly introduced ene-type cyclization reaction was used to prepare a series of epiminopyrimido[4,5-b]azepines. To gain insight into the mechanism of the two reaction pathways, a DFT study was carried out. Theoretical calculations showed that under acidic conditions an iminium intermediate favors the low-energy IADA pathway, which proceeds in a [4⁺ + 2] fashion. When acid is absent, the neutral imine intermediate favors the thermal ene-type cyclization reaction, which takes place by transfer of an allylic proton from the allylic amine to imine, followed by a barrierless nucleophilic addition process between the in-situ-generated anionic allylic amine and iminium ion. Amine addition to the alkene finally gives the epiminopyrimido[4,5-b]azepines. A complete switch from an intramolecular aza-Diels-Alder (IADA) reaction to an ene-type cyclization was achieved simply by adjusting the reaction conditions. A DFT study was carried out to gain insight into the mechanisms of the two reaction pathways. A series of epiminopyrimido[4,5-b]azepines were prepared using this newly introduced ene-type cyclization reaction.

Full text of DOI:10.1002/ejoc.201301318

FULL PAPER
DOI: 10.1002/ejoc.201301318
On-Demand Selection of the Reaction Path from Imino Diels–Alder to Ene-
Type Cyclization: Synthesis of Epiminopyrimido[4,5-b]azepines
Yuewei Zhang,[a] Yue Zhu,[a] Lianyou Zheng,[a] Lian-Gang Zhuo,[b] Fengzhi Yang,[a]
Qun Dang,[a] Zhi-Xiang Yu,*[b] and Xu Bai*[a]
Keywords: Synthetic methods / Nitrogen heterocycles / Fused-ring systems / Cyclization / Ene reaction / Density functional
calculations
Controlling the mode of reaction of a reactive intermediate
such as an imine or iminium ion should enable the on-de-
mand selection of the final products from the same starting
materials. The successful execution of such a strategy will
reduce the time required to prepare diverse scaffolds. The
minopyrimido[4,5-b]azepines. To gain insight into the
mechanism of the two reaction pathways, a DFT study was
carried out. Theoretical calculations showed that under
acidic conditions an iminium intermediate favors the low-en-
ergy IADA pathway, which proceeds in a [4+ + 2] fashion.
imines derived from 4-(allylamino)pyrimidine-5-carbal- When acid is absent, the neutral imine intermediate favors
dehydes and anilines undergo Diels–Alder reactions to give the thermal ene-type cyclization reaction, which takes place
pyrimido[4,5-h][1,6]naphthyridines in high yields. A com- by transfer of an allylic proton from the allylic amine to imine,
plete switch from the intramolecular aza-Diels–Alder (IADA)
path to an ene-type cyclization reaction was achieved by
simply adjusting the reaction conditions (amount of acid cat-
alyst, solvent, and temperature). This newly introduced ene-
type cyclization reaction was used to prepare a series of epi-
followed by a barrierless nucleophilic addition process be-
tween the in-situ-generated anionic allylic amine and imin-
ium ion. Amine addition to the alkene finally gives the ep-
iminopyrimido[4,5-b]azepines.
heterocycles such as alkaloids, and they also represent two
alternative reaction pathways available to suitable imines.
They share several features: both involve six electrons, and
their mechanisms may be pericyclic or stepwise. Thus, they
may exist as competing reactions. However, the activation
energy of a simple alkene ene reaction is generally higher
than that of the IADA reaction, so it usually requires higher
temperatures, and this may limit its synthetic application.[5]
Recently, the discovery of Lewis-acid- and Brønsted-acid-
catalyzed ene reactions of a variety of substrates has ex-
panded their utility in organic synthesis.[5b,6] We wondered
whether these two reactions could be used to generate dif-
ferent scaffolds from the same starting materials at will by
changing the reaction conditions. Noguchi’s group has
thoroughly investigated the intramolecular ene-like reac-
tions of several heterocyclic substrates,[7] and they reported
that a competition exists between thermal ene reactions and
IADA reactions in 2-[N-(alk-2-enyl)benzylamino]-3-vinyl-
pyrido[1,2a]pyrimidin-4(4H)-ones.[7k] In addition, Nagara-
jan et al. have studied the switch from IADA to ene-type
cyclization of aminoanthraquinones with citronellal or
prenylated salicylaldehydes resulting from changes in the
substituents.[8] However, to date, a complete switch from an
IADA to an ene-type reaction of an imine involving the
same set of reactants has not been reported.
Introduction
Efficient access to diverse scaffolds is highly desirable in
organic and medicinal chemistry.[1] Different scaffolds usu-
ally require different sets of starting materials, which must
be prepared separately, leading to additional effort and cost.
However, organic compounds very often contain multiple
reactive sites that can potentially lead to multiple products
by following different pathways. It is therefore desirable for
different scaffolds to be prepared from the same set of start-
ing materials by simply changing the reaction conditions.
This has been an important area of research in synthetic
organic chemistry.[2]
The intramolecular inverse electron-demand aza-Diels–
Alder (IADA) reaction (also known as the Povarov reac-
tion[3]) and the ene reaction[4] are two powerful methods
that can be used for the synthesis of nitrogen-containing
[a] The Center for Combinatorial Chemistry and Drug Discovery,
The College of Chemistry and The School of Pharmaceutical
Sciences, Jilin University,
1266 Fujin Lu, Changchun Jilin 130021, P. R. China
E-mail: xbai@jlu.edu.cn
http://juccc.jlu.edu.cn
[b] College of Chemistry, Peking University,
Beijing 100871, P. R. China
E-mail: yuzx@pku.edu.cn
http://www.chem.pku.edu.cn/page/zxyu/
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.201301318.
As part of our ongoing research into the synthesis of
heterocyclic scaffolds,[9] we previously prepared hexa-
660
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 660–669

Switching between Imino Diels–Alder and Ene-Type Cyclization
Scheme 1. The IADA and ene-type cyclization pathways of imine 4a.
hydrobenzo[b]pyrimido[4,5-h][1,6]naphthyridines 2a by the
IADA reaction of imine intermediate 4a under acidic condi-
tions (for details, see the left part of Figure 1b and the cor-
responding discussions in the DFT calculation section;
path a in Scheme 1).[10] Subsequently, we recognized that
imine 4a might also undergo an ene-type cyclization to pro-
Results and Discussion
Transition and Switch between the IADA and Ene-Type
Reactions
Previously, we reported that treatment of (allylamino)-
duce another fused heterocycle 3a (for details, see the right pyrimidinecarbaldehyde 1a with aniline in the presence of
part of Figure 1a and the corresponding discussions in the trifluoroacetic acid (TFA) in acetonitrile/water at ambient
DFT calculation section), provided suitable reaction condi- temperature generated cis-configured tetracyclic IADA re-
tions could be identified. In this aspect, we have thoroughly action product 2a (path a, Scheme 1; Table 1, Entry 1).[10]
investigated the transitions between IADA and ene-type cy- Subsequently, we recognized that imine intermediate 4a had
clizations in this unique molecular system under various all the elements required for an intramolecular ene-type re-
conditions, and we have obtained a complete switch from action, and that – if successfully executed – this ene-type
the IADA to the ene-type reaction product in nearly quanti- cyclization could be developed into a useful method to pre-
tative yield. In this paper, the details of this conditions-ori- pare heterocycles such as compounds 3a (path b,
ented synthesis are discussed.
Scheme 1). To direct the imine reaction towards the ene cy-
Table 1. Investigation of IADA vs. ene-type reactions.[a]
Entry
Solvent
Catalyst (equiv.)
Temp [°C]
Time [h]
Ratio[b] 2a/3a
Yield [%][c]
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
MeCN/H2O (1:1)
DMF
TFA (2.0)
TsOH (1.0)
TsOH (1.0)
TsOH (1.0)
TsOH (1.0)
TsOH (1.0)
TsOH (0.05)
TsOH (0.05)
TsOH (0.05)
TsOH (0.05)
TsOH (0.05)
TsOH (0.05)
none
25
11
1
1
1
1
0.5
6
3
1.5
16
3
3.5
24
15
30
48
6.5
48
48
2a
2a
94
71
86
78
94
95
83
77
70
115
115
115
DMSO
nBuOH
benzene
toluene
nBuOH
DMF
DMSO
benzene
toluene
toluene
MeCN
benzene
nBuOH
DMF
DMSO
toluene
toluene
2a
2a
[d]
80
2a
[d]
111
2a
115
115
115
2.1:1
2.3:1
2.1:1
3a[e]
3a[e]
3a[e]
–
[d]
80
93
97
95
[d]
111
111
81
0[f]
0[f]
46[g]
51
[d]
none
none
none
none
none
none
80
–
3a
115
115
115
111
3a[h]
3a[e]
3a[h]
3a[h]
67
28[i]
57[j]
[d]
111
[a] Unless otherwise noted, reactions were carried out with 1a (0.50 mmol) and aniline (0.525 mmol) in solvent (4 mL). [b] Determined
from the 1H NMR spectrum of the crude reaction mixture. [c] Isolated yield after flash column chromatography. [d] Using a Dean–Stark
trap to remove water. [e] A trace of 2a was detected, but it could not be isolated. [f] Starting material 1a was recovered (98%). [g] Starting
material 1a was recovered (17%). [h] Contained carbonyl-ene reaction product. [i] Starting material 1a was recovered (63%). [j] Starting
material 1a was recovered (29%).
Eur. J. Org. Chem. 2014, 660–669
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
661

Z.-X. Yu, X. Bai et al.
FULL PAPER
clization pathway, we decided to investigate reaction condi- state energy barrier, since the reaction only occurred at
tions such as solvent, temperature, and acid catalyst, and to higher temperatures. The IADA reaction proceeded at
delineate their influences on the two competing reactions. much lower temperatures.
This could guide us to select the best conditions for a de-
sired reaction pathway. The results of this investigation are
summarized in Table 1.
Experimental Rationalization of the Transition from the
IADA to the Ene-Type Reaction
In our previous studies, we had found that TFA was an
important component for the IADA reaction;[10] so to steer
the reaction away from the IADA pathway, we first investi-
The above results can readily be rationalized by the exis-
gated the effects of TFA on the two reaction pathways. Re- tence of two competing reaction processes, as shown in
placing TFA with p-toluenesulfonic acid (TsOH; 1 equiv.) Scheme 2. The reaction of pyrimidinecarbaldehyde 1a with
and exploring solvents with various polarities led only to aniline should initially form imine intermediate 4a, which
the IADA reaction product (i.e., 2a; Table 1, Entries 2–6). may undergo a thermal ene-type reaction (higher activation
Next the amount of TsOH was reduced from a full equiva- energy) directly to yield product 3a,[7f] or proceed through
lent to a catalytic amount, and encouragingly, we began to an IADA reaction (lower activation energy) via iminium
see a mixture of IADA and ene-type reaction products. We ion 5a to generate product 2a.[10] The formation of the
noted that the IADA product (i.e., 2a) was predominant phenyliminium ion is required for the low-energy IADA re-
when a polar solvent was used (Table 1, Entries 7–9), action in which the phenyliminium ion acts as the azadiene.
whereas only the ene product (i.e., 3a) was obtained in high When the same reaction was carried out under thermal con-
yield when the reaction was conducted in a non-polar sol- ditions in the absence of acid, only the ene-type cyclization
vent such as benzene or toluene (with water removal using product was formed. Therefore, a polar solvent, low tem-
a Dean–Stark trap; Table 1, Entries 10 and 11). To further perature, and the presence of a proton source favor the
investigate the acid issue, we decided to test the reaction in IADA reaction pathway. This analysis is consistent with the
the absence of any acid (Table 1, Entries 13–19). When the experimental results shown in Table 1. A catalytic amount
reaction was carried out at 81 °C in the absence of acid, of acid promoted the formation of imine intermediate 4a.
neither product was obtained, regardless of whether a polar This intermediate either continued on the ene-type pathway
(Table 1, Entry 13) or a non-polar (Table 1, Entry 14) sol- to give 3a (Table 1, Entry 11), as was seen in the non-polar
vent was used. When the reaction temperature was in- solvent toluene at high temperature, or, in a polar solvent,
creased to 111 °C (Table 1, Entries 15–19) the ene-type reac- it resulted in a mixture of 2a and 3a (Table 1, Entries 7–9).
tions were facilitated, although they required longer reac- This is seen, because in a non-polar solvent, the formation
tion times and led to lower yields compared to Table 1, En- of an imine intermediate is favored over the iminium ion,
tries 10–12. These results indicate that catalytic amounts of whereas in the polar solvent, the iminium intermediate is
acid and less polar solvents favor the ene-type reaction, favored. In the absence of an acid catalyst and at elevated
which is consistent with the envisioned reaction mecha- temperature, pyrimidinecarbaldehyde 1a could either react
nisms.
with aniline to form imine 4a, which could undergo an ene-
Therefore, we were able to induce a complete switch from type reaction to give product 3a, or it could overcome a
the previously reported IADA reaction[10] to an ene-like re- higher energy barrier to give the carbonyl-ene product
action with the same set of starting materials in nearly (Table 1, Entries 15, 16, 18, and 19). In refluxing aceto-
quantitative yields by simply changing the reaction condi- nitrile or benzene, i.e., low-boiling solvents, in the absence
tions. The ene-type reaction seems to have a high transition- of an acid catalyst (Table 1, Entries 13 and 14), no imine
Scheme 2. Two competing reaction processes.
662
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 660–669

Switching between Imino Diels–Alder and Ene-Type Cyclization
Scheme 3. Verification of the iminium ion leading to the IADA product.
or iminium intermediate was formed, so no product was energy surface (PES) scan (see Supporting Information for
obtained from either reaction pathway.
details) found that TS2 connects 4a and IN2. The calcula-
To further verify the above proposed processes, pyrimid- tions did not find any intermediates between TS2 and IN2,
inecarbaldehyde 1a and N-methylaniline were subjected to neither in the gas phase nor in toluene solution.[15] This
the optimized reaction conditions of the ene-like reaction suggests that the ene-type reaction can be regarded as a
[i.e., TsOH (0.05 equiv.), refluxing toluene with removal of proton transfer from the allylic moiety to the enamine, gen-
water; Scheme 3].[10] As expected, only the IADA product erating a zwitterionic intermediate, IN3, which has an imin-
was obtained. This is, because only the IADA pathway im- ium ion moiety and a carbanion moiety (see Figure 1a).[16]
inium intermediate (i.e., 5b) should be formed in this case. Intermediate IN3 does not represent an energy minimum,
This experiment provided additional evidence that the neither in the gas phase nor in solution, and it can easily
IADA product was derived from an iminium intermediate. collapse (in a barrierless process) to give IN2 by nucleo-
Following on from this, DFT calculations were used to gain philic addition. Due to the highly asynchronous character
further mechanistic insight.
of TS2, it is better to describe this step as a [1,6]-proton
shift, followed by a rapid nucleophilic addition. Intermedi-
ate IN2 then can be converted into the final tertiary amine
(i.e., 3a) by addition of the secondary amine to the C=C
double bond, which could be catalyzed by a trace amount
DFT Study of the Mechanisms and Origins of
Regioselectivity
To gain further insight into the mechanisms and the ori- of Brønsted acid.[17]
gins of the experimentally observed regioselectivity, theoret-
However, for protonated species 5a (Figure 1b), we could
ical calculations were carried out to investigate both the not locate an ene-like transition state (HTS2). This was at-
IADA and ene-type reaction pathways, using imine 4a or tributed to the fact that the nitrogen atom in 5a is an am-
iminium ion 5a as the model reactants (Figure 1). All calcu- monium ion, which prevents any proton transfer (instead,
lations were carried out with Gaussian 03 programs.[11] there is a possible [1,5]-hydride shift process via transition
Geometrical optimizations of all species were performed at state HTS3, but this is not favored either). In contrast, the
the B3LYP/6-311+G(d,p) level.[12] Solvent effects were com- IADA reaction of 5a, which can be termed a [4+ + 2][18]
puted using UAHF (united atom topological model for reaction between the protonated phenylimine moiety and
Hartree–Fock) radii and a CPCM (conductor polarized the terminal olefin moiety, can take place by either of two
continuum model)[13] solvent model in toluene. Unless competing pathways, the endo- and the exo-IADA path-
otherwise specified, all the energies discussed in this paper ways. The endo-IADA transition state (i.e., endo-HTS1),
are the Gibbs free energies in toluene at 298 K (ΔGsol 298). which gives rise to intermediate cis-HIN1, is more favored
Gibbs free energies in the gas phase (ΔGgas 298) and gas than the exo transition state (i.e., exo-HTS1), which gives
phase enthalpies (ΔH) are also provided for reference.
rise to intermediate trans-HIN1, by 4.9 kcal/mol.[19] The
DFT calculations found that imine 4a can be converted IADA reactions are irreversible, since the reverse reactions
into IN2 via TS2 with an activation Gibbs free energy in are difficult with activation energies of Ͼ30 kcal/mol. After
toluene of 24.1 kcal/mol in the ene-type pathway (Fig- the IADA reaction, cycloadducts cis-HIN1 and trans-HIN1
ure 1a). Transition state TS2 is more favored than the com- can then be transformed into cis-2a and trans-2a by depro-
peting IADA transition states, endo-TS1 and exo-TS1, by tonation, which is not expected to be the rate-limiting step.
3.5 and 8.7 kcal/mol, respectively. These calculations sug- The results of these calculations suggest that the cis Diels–
gested that the ene-type reaction is favored over the IADA Alder product (i.e., cis-2a) should be obtained as the major
reactions, which is consistent with the experimental obser- product from 5a, which matches the experimental results in
vation that in the absence of a proton source, ene-type Table 1.
product 3a was obtained exclusively. It is interesting to find
The activation energy of the [4+ + 2] transition state (i.e.,
that in the ene-type reaction transition state (i.e., TS2), the endo-HTS1; Figure 1b) is 16.0 kcal/mol, which is much
allylic proton, which is also adjacent to the nitrogen atom, lower than that of the [4 + 2] transition state (i.e., endo-
is transferring to the enamine moiety, while the C-1–C-4 TS1, 27.6 kcal/mol; Figure 1a). This observation can be ra-
bond formation is not taking place at all, with the distance tionalized using frontier molecular orbital (FMO) theory.
between C-1 and C-4 being 3.36 Å (Figure 1c). IRC (intrin- Protonated 5a has a significantly lower-energy LUMO,
sic reaction coordinate)[14] calculations and the potential which is mainly composed of the diene (phenylimine moi-
Eur. J. Org. Chem. 2014, 660–669
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
663

Z.-X. Yu, X. Bai et al.
FULL PAPER
Figure 1. (a) Two calculated reaction pathways of 4a. (b) Two calculated reaction pathways of 5a. (c) Selected DFT calculated structures
(distances in Å).
ety) π orbitals. The reacting π orbital of the dienophile is
A comparison of the results of the calculations shown in
HOMO-6 (Figure 2). The energy gap between the LUMO Figure 1a and b reveals that in the presence of a full equiva-
and HOMO-6 is 7.5 eV. In contrast, for 4a, the correspond- lent of acid, the phenylimine moiety of 5a is protonated,
ing energy gap between the two reacting orbitals is 8.3 eV, which leads to a lower LUMO energy and favors the IADA
which is 0.8 eV larger than that in 5a. Therefore, the IADA reaction. In contrast, the ene-type reaction cannot occur,
reaction between the protonated phenylimine and the “elec- since the protonated amine moiety cannot accept a proton
tron-rich” olefin in 5a become favored.
from the allylic moiety in 5a. When an acid is absent, or
664
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 660–669

Switching between Imino Diels–Alder and Ene-Type Cyclization
various functional groups, such as nitro, ester, cyano, and
halo groups, were tolerated, producing good to excellent
yields of epiminopyrimido[4,5-b]azepines. However, when
an ortho-nitro group was present (Table 2, Entry 4), a lower
yield was observed, which may be due to a steric effect.
Next, the scope of the R1 group was investigated. Halo,
alkoxy, phenoxy, and amino groups were found to be suit-
able substituents, and moderate to good yields were ob-
tained (Table 2, Entries 10–13).
Figure 2. FMOs of 4a and 5a, orbital energies calculated at the
HF/6-31G(d) level.
Conclusions
only a catalytic amount of acid is present, the IADA reac-
tion under neutral conditions is difficult, due to its higher
LUMO energy. On the other hand, an ene-type reaction
can take place, due to the easier [1,6]-proton transfer and
barrierless nucleophilic addition.
A complete switch from an imine Diels–Alder reaction
to an ene-type cyclization reaction was achieved. The same
set of starting materials could be used to prepare two dis-
tinct heterocyclic scaffolds simply by changing the reaction
conditions, such as the amount of acid catalyst, the solvent,
and the reaction temperature. This approach avoids the us-
ual practice of using different sets of starting materials for
the synthesis of different scaffolds, and therefore makes ac-
Synthesis of Epiminopyrimido[4,5-b]azepines
Having identified the optimal conditions for the ene-type cess of new scaffolds more efficient. These results could be
reaction pathway, the scope of the reaction was explored rationalized by the existence of two competing processes:
for the preparation of epiminopyrimido[4,5-b]azepines. the low-activation-energy Diels–Alder cyclization with an
Thus, the reaction was tested with various pyrimidinecarb- iminium ion as the intermediate, and the thermal ene-type
aldehydes and amines, using a catalytic amount of TsOH, cyclization with an imine as the key intermediate. To gain
refluxing toluene, and a Dean–Stark trap for the removal insight into the mechanisms of the two reaction pathways,
of water. The results are summarized in Table 2.
a DFT study was carried out. Theoretical calculations
Generally, this ene-type reaction has a broad scope, and showed that under acidic conditions, an iminium intermedi-
variations of the R1, R2, and X groups were tolerated to ate favors the low-energy IADA pathway, while the neutral
give epiminopyrimido[4,5-b]azepines in good yields. First, imine intermediate favors the thermal ene-type cyclization
we explored the scope of the R2 group with various amines pathway as the low-energy process. The ene-type cyclization
(Table 2, entries 1–8). Both aryl and aliphatic R2 groups led was further successfully applied to the preparation of a
to productive ene-type products, with aryl groups giving series of epiminopyrimido[4,5-b]azepines. These results
higher yields than aliphatic groups. Aromatic amines with could serve as an interesting example of efficiency in or-
ganic synthesis and green chemistry, and could encourage
further studies in the generation of diverse scaffolds from
the same sets of starting materials.
Table 2. Selected imine ene-type reactions.
Experimental Section
General Methods: Unless otherwise noted, reactions were carried
out in air. Chemicals and solvents were purchased from commercial
suppliers, and were either used as supplied or purified by standard
procedures. Analytical thin-layer chromatography (TLC) was per-
Entry
1
R1
X
R2
Time Product
[h]
Yield
[%]
formed on silica gel plates with F-254 indicator, and compounds
were visualized by irradiation with UV light or by treatment with
iodine. Flash column chromatography was performed with silica
gel 200–300 mesh. Mass spectra and HPLC data were recorded
with an LC–MS system with ELSD (evaporative light-scattering
detection). 1H and 13C NMR spectra were recorded at 300 and
75 MHz, respectively. The spectra were recorded in CDCl3 at room
temperature unless otherwise noted. 1H and 13C NMR chemical
shifts are reported in ppm relative to TMS, which was used as an
internal standard. Multiplicities are indicated as follows: s, singlet;
d, doublet; t, triplet; q, quartet; m, multiplet; dd, doublet of dou-
blets; br., broad. Compounds 1a, 1c, 2a, 2b, 2e are known com-
pounds, and were prepared according to the previous report.[10]
1
2
3
4
5
6
7
8
9
10
11
12
13
1a
1a
1a
1a
1a
1a
1a
1a
1b
1c
PhS NMe
PhS NMe
PhS NMe
PhS NMe
Ph
3.5
4.5
5
7
2.5
6
3a
3b
3c
3d
3e
3f
97
93
95
66
88[a]
60
61
84
95
80
67
70
44
4-NCC6H4
4-O2NC6H4
2-O2NC6H4
PhS NMe 4-MeOC6H4
PhS NMe
PhS NMe
PhS NMe
PhS NAllyl
EtO2CCH2
nBu
Bn
Ph
Ph
Ph
Ph
Ph
86
10
1.5
2
3g
3h
3i
Cl
NMe
3j
1d PhO NMe
1e MeO NMe
1f Et2N NMe
5.5
5
3k
3l
23
3m
4-[Allyl(methyl)amino]-6-(phenylthio)pyrimidine-5-carbaldehyde
[a] Diels–Alder product 2e was isolated (6%).
(1a): Pale yellow solid (0.541 g, 95%). M.p. 57–58 °C. 1H NMR
Eur. J. Org. Chem. 2014, 660–669
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
665

Z.-X. Yu, X. Bai et al.
FULL PAPER
(300 MHz, CDCl3): δ = 10.23 (s, 1 H, CHO), 8.24 (s, 1 H, pyr-
4-[Allyl(methyl)amino]-6-methoxypyrimidine-5-carbaldehyde (1e):
imidyl-H), 7.57–7.52 (m, 2 H, phenyl-H), 7.46–7.43 (m, 3 H, Potassium carbonate (0.276 g, 2.0 mmol) was added to a solution
phenyl-H), 5.94–5.85 (m, 1 H, CH=C), 5.32–5.23 (m, 2 H, of 1c (0.211 g, 1.0 mmol) in MeOH (2 mL). The reaction mixture
C=CH2), 4.23 (d, J = 5.1 Hz, 2 H, CH2), 3.10 (s, 3 H, CH3) ppm. was heated at reflux for 4 h. The reaction solution was then diluted
13C NMR (75 MHz, CDCl3): δ = 186.7, 173.4, 163.0, 157.0, 135.4, with water (10 mL) and extracted with EtOAc (3ϫ 15 mL). The
132.3, 129.3, 129.0, 128.9, 118.4, 109.4, 55.3, 39.6 ppm. MS (ESI):
combined organic extracts were washed with brine, dried with an-
hydrous Na2SO4, and concentrated in vacuo. Purification by flash
column chromatography (petroleum ether/EtOAc, 5:1, v/v) gave 1e
m/z = 285.9 [M + H]+.
4-(Diallylamino)-6-(phenylthio)pyrimidine-5-carbaldehyde (1b): A
mixture of Et3N (1.220 mL, 8.8 mmol) and diallylamine (1.080 mL,
8.8 mmol) was slowly added to a solution of 4,6-dichloro-5-formyl-
pyrimidine (1.408 g, 8.0 mmol) in anhydrous CHCl3 (12 mL) at
0 °C. The reaction mixture was stirred at ambient temperature for
30 min, then the mixture was quenched with water (15 mL) and
extracted with CH2Cl2 (3ϫ 15 mL). The combined organic extracts
were washed with brine, dried with anhydrous Na2SO4, and con-
centrated in vacuo. Purification by flash column chromatography
(petroleum ether/EtOAc, 5:1, v/v) gave 4-chloro-6-(diallylamino)-
pyrimidine-5-carbaldehyde (1.855 g, 95%) as a white solid. M.p.
44–45 °C. Thiophenol (0.220 g, 2.0 mmol) was added to a solution
of 4-chloro-6-(diallylamino)pyrimidine-5-carbaldehyde (0.475 g,
2.0 mmol) and Et3N (0.270 mL, 2.0 mmol) in CHCl3 (4 mL) at am-
bient temperature. The reaction mixture was stirred for 1.5 h, then
it was diluted with water (10 mL) and extracted with CH2Cl2 (3ϫ
10 mL). The combined organic extracts were washed with brine,
dried with anhydrous Na2SO4, and concentrated in vacuo. Purifica-
tion by flash column chromatography (petroleum ether/EtOAc, 5:1,
v/v) gave 1b (0.520 g, 84 %) as a pale yellow oil. 1H NMR
(300 MHz, CDCl3): δ = 10.13 (s, 1 H, CHO), 8.25 (s, 1 H, pyrim-
idyl-H), 7.54 (dd, J = 6.6, 3.0 Hz, 2 H, phenyl-H), 7.44 (m, 3 H,
phenyl-H), 5.95–5.82 (m, 2 H, CH=C), 5.32 (d, J = 1.2 Hz, 1 H,
C=CH2), 5.28 (dd, J = 1.8, 1.2 Hz, 2 H, C=CH2), 5.22 (d, J =
1.2 Hz, 1 H, C=CH2), 4.16 (d, J = 5.4 Hz, 4 H, 2 CH2) ppm. 13C
NMR (75 MHz, CDCl3): δ = 186.9, 172.9, 163.4, 157.0, 135.4,
132.5, 129.2, 129.1, 128.8, 118.6, 109.9, 52.9 ppm. MS (ESI): m/z
= 311.9 [M + H]+. HRMS (ESI-TOF): calcd. for C17H18N3OS [M
+ H]+ 312.1157; found 312.1165.
1
(0.151 g, 73%) as a pale yellow oil. H NMR (300 MHz, CDCl3):
δ = 10.24 (s, 1 H, CHO), 8.27 (s, 1 H, pyrimidyl-H), 5.95–5.82 (m,
1 H, CH=C), 5.25–5.17 (m, 2 H, C=CH2), 4.41–4.15 (m, 2 H,
CH2), 4.04 (s, 3 H, OCH3), 2.92 (s, 3 H, CH3) ppm. 13C NMR
(75 MHz, CDCl3): δ = 186.4, 172.5, 161.7, 158.2, 132.6, 117.8, 99.3,
55.4, 54.3, 39.6 ppm. MS (ESI): m/z = 208.3 [M + H]+. HRMS
(ESI-TOF): calcd. for C10H14N3O2 [M + H]+ 208.1078; found
208.1081.
4-[Allyl(methyl)amino]-6-(diethylamino)pyrimidine-5-carbaldehyde
(1f): Diethylamine (0.146 g, 2.0 mmol) was added to a solution of
1c (0.211 g, 1.0 mmol) and Et3N (0.151 g, 1.5 mmol) in CHCl3
(2 mL). The reaction mixture was heated at reflux for 2 h. The mix-
ture was then diluted with water (10 mL) and extracted with EtOAc
(3 ϫ 15 mL). The combined organic extracts were washed with
brine, dried with anhydrous Na2SO4, and concentrated in vacuo.
Purification by flash column chromatography (petroleum ether/
EtOAc, 2:1, v/v) gave 1f (0.161 g, 65%) as a colorless oil. 1H NMR
(300 MHz, CDCl3): δ = 9.37 (s, 1 H, CHO), 8.06 (s, 1 H, pyrimidyl-
H), 5.96–5.85 (m, 1 H, CH=C), 5.29–5.26 (m, 1 H, C=CH2), 5.23–
5.22 (m, 1 H, C=CH2), 4.27 (dt, J = 5.7, 1.5 Hz, 2 H, CH2), 3.69
(q, J = 7.2 Hz, 4 H, 2 CH2), 3.15 (s, 3 H, CH3), 1.29 (t, J = 7.2 Hz,
6 H, 2 CH3) ppm. 13C NMR (75 MHz, CDCl3): δ = 181.5, 165.9,
164.9, 158.1, 133.0, 117.8, 96.0, 54.7, 44.6, 39.1, 13.3 ppm. MS
(ESI): m/z = 249.4 [M + H]+. HRMS (ESI-TOF): calcd. for
C13H21N4O [M + H]+ 249.1697; found 249.1710.
5-Methyl-1-(phenylthio)-5,6,6a,7,12,12a-hexahydrobenzo[b]pyr-
imido[4,5-h][1,6]naphthyridine (2a): White solid (0.169 g, 94%).
1
M.p. 203–204 °C. H NMR (300 MHz, CDCl3): δ = 8.28 (s, 1 H,
4-[Allyl(methyl)amino]-6-chloropyrimidine-5-carbaldehyde (1c):
White solid (0.400 g, 96%). M.p. 53–54 °C. 1H NMR (300 MHz,
CDCl3): δ = 10.38 (s, 1 H, CHO), 8.35 (s, 1 H, pyrimidyl-H), 5.86–
5.79 (m, 1 H, CH=C), 5.25–5.19 (m, 2 H, C=CH2), 4.31–4.29 (m,
2 H, CH2), 2.89 (s, 3 H, CH3) ppm. MS (ESI): m/z = 212.0 [M +
H]+.
pyrimidyl-H), 7.56–7.52 (m, 2 H, phenyl-H), 7.43–7.39 (m, 3 H,
phenyl-H), 7.06–7.00 (m, 2 H, phenyl-H), 6.68 (t, J = 7.2 Hz, 1 H,
phenyl-H), 6.51 (d, J = 7.8 Hz, 1 H, phenyl-H), 4.75 (d, J = 2.4 Hz,
1 H, CH), 3.92 (s, 1 H, NH), 3.56 (t, J = 12.3 Hz, 1 H, CH2), 3.30
(dd, J = 17.1, 6.3 Hz, 1 H, CH2), 3.16 (s, 3 H, CH3), 3.12 (dd, J =
12.6, 4.8 Hz, 1 H, CH2), 2.62 (d, J = 17.1 Hz, 1 H, CH2), 2.47 (m,
1 H, CH) ppm. 13C NMR (75 MHz, CDCl3): δ = 162.9, 157.7,
157.3, 141.5, 134.8, 129.7, 129.4, 129.0, 127.6, 118.2, 118.1, 114.8,
111.3, 90.6, 49.7, 47.5, 36.3, 29.2, 28.8 ppm. MS (ESI): m/z = 361.1
[M + H]+.
4-[Allyl(methyl)amino]-6-phenoxypyrimidine-5-carbaldehyde (1d):
Sodium hydride (0.080 g, 2.0 mmol) was added in portions to a
solution of phenol (0.094 g, 1.0 mmol) in THF (3 mL) in an ice/
water bath under nitrogen. The mixture was stirred for 30 min, then
a solution of compound 1c (0.211 g, 1.0 mmol) in THF (1 mL) was
added dropwise. The reaction mixture was stirred in the ice/water
bath for a further 2 h. Then the mixture was diluted with water
(15 mL) and extracted with EtOAc (3ϫ 15 mL). The combined or-
ganic extracts were washed with brine, dried with anhydrous
Na2SO4, and concentrated in vacuo. Purification by flash column
chromatography (petroleum ether/EtOAc, 1:1, v/v) gave 1d
(0.224 g, 83%) as a colorless oil. 1H NMR (300 MHz, CDCl3): δ
= 10.47 (s, 1 H, CHO), 8.22 (s, 1 H, pyrimidyl-H), 7.47–7.42 (m, 2
H, phenyl-H), 7.31–7.28 (m, 1 H, phenyl-H), 7.20–7.13 (m, 2 H,
phenyl-H), 6.00–5.81 (m, 1 H, CH=C), 5.28–5.27 (m, 1 H,
C=CH2), 5.26–5.21 (m, 1 H, C=CH2), 4.30 (dt, J = 5.7, 1.5 Hz, 2
H, CH2), 2.98 (s, 3 H, CH3) ppm. 13C NMR (75 MHz, CDCl3): δ
= 186.5, 172.4, 162.1, 158.4, 152.5, 132.5, 129.6, 125.7, 121.7, 118.2,
99.9, 54.6, 40.0 ppm. MS (ESI): m/z = 270.4 [M + H]+ .
C15H15N3O2 (269.30): calcd. C 66.90, H 5.61, N 15.60; found C
66.69, H 5.36, N 15.65.
5,12-Dimethyl-1-(phenylthio)-5,6,6a,7,12,12a-hexahydrobenzo[b]pyr-
imido[4,5-h][1,6]naphthyridine (2b): White solid (0.134 g, 78 %).
1
M.p. 167–169 °C. H NMR (300 MHz, CDCl3): δ = 8.27 (s, 1 H,
pyrimidyl-H), 7.54–7.52 (m, 2 H, phenyl-H), 7.40–7.39 (m, 3 H,
phenyl-H), 7.17 (t, J = 7.2 Hz, 1 H, phenyl-H), 7.03 (d, J = 6.6 Hz,
1 H, phenyl-H), 6.76–6.73 (m, 2 H, phenyl-H), 4.60 (s, 1 H, CH),
3.57 (t, J = 12.6 Hz, 1 H, CH2), 3.33 (dd, J = 17.1, 6.6 Hz, 1 H,
CH2), 3.14 (s, 3 H, CH3), 3.14–3.09 (m, 1 H, CH2), 2.84 (s, 3 H,
CH3), 2.66 (d, J = 17.1 Hz, 1 H, CH2), 2.48 (br. s, 1 H, CH) ppm.
13C NMR (75 MHz, CDCl3): δ = 164.7, 157.0, 144.9, 134.9, 129.5,
129.0, 128.8, 127.6, 119.7, 117.4, 112.7, 109.4, 53.1, 49.5, 35.8, 34.0,
29.7, 29.2 ppm. MS (ESI): m/z = 375.2 [M + H]+.
9-Methoxy-5-methyl-1-(phenylthio)-5,6,6a,7,12,12a-hexahydro-
benzo[b]pyrimido[4,5-h][1,6]naphthyridine (2e): White solid (0.012 g,
1
6%). M.p. 201–202 °C. H NMR (300 MHz, CDCl3): δ = 8.28 (s,
1 H, pyrimidyl-H), 7.53–7.51 (m, 2 H, phenyl-H), 7.39–7.37 (m, 3
666
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 660–669

Switching between Imino Diels–Alder and Ene-Type Cyclization
H, phenyl-H), 6.66 (d, J = 8.1 Hz, 1 H, phenyl-H), 6.60 (s, 1 H, 128.8, 126.0, 121.7, 121.5, 114.1, 75.7, 57.7, 34.5, 33.7, 33.0 ppm.
phenyl-H), 6.48 (d, J = 8.1 Hz, 1 H, phenyl-H), 4.69 (s, 1 H, CH), MS (ESI): m/z = 406.1 [M + H]+. C21H19N5O2S (405.47): calcd. C
3.82 (s, 4 H, OCH3, NH), 3.58 (t, J = 12.3 Hz, 1 H, CH2), 3.30 62.21, H 4.72, N 17.27, S 7.91; found C 62.00, H 4.63, N 17.33, S
(dd, J = 16.8, 5.7 Hz, 1 H, CH2), 3.16 (s, 3 H, CH3), 3.11 (dd, J =
12.3, 3.6 Hz, 1 H, CH2), 2.59 (d, J = 17.7 Hz, 1 H, CH2), 2.45 (br.
s, 1 H, CH) ppm. 13C NMR (75 MHz, CDCl3): δ = 162.5, 157.5,
157.1, 152.2, 135.2, 134.5, 129.4, 129.0, 128.6, 119.0, 115.6, 114.4,
113.4, 111.1, 55.6, 49.3, 47.4, 36.0, 29.1, 28.5 ppm. MS (ESI): m/z
= 391.1 [M + H]+.
7.80.
(5R*,8R*)-10-(4-Methoxyphenyl)-9-methyl-4-(phenylthio)-6,7,8,9-
tetrahydro-5H-5,8-epiminopyrimido[4,5-b]azepine (3e): White solid
(0.172 g, 88%). M.p. 190–192 °C. H NMR (300 MHz, CDCl3): δ
1
= 8.13 (s, 1 H, pyrimidyl-H), 7.50–7.48 (m, 2 H, phenyl-H), 7.37–
7.36 (m, 3 H, phenyl-H), 6.83–6.75 (m, 4 H, phenyl-H), 5.11–5.09
(m, 2 H, 2 CH), 3.73 (s, 3 H, OCH3), 3.12 (s, 3 H, CH3), 2.48–2.23
(m, 3 H, CH2), 2.12–2.06 (m, 1 H, CH2) ppm. 13C NMR (75 MHz,
CDCl3): δ = 159.1, 156.7, 156.4, 153.6, 138.5, 133.9, 130.3, 129.1,
128.4, 118.5, 115.6, 114.7, 75.1, 56.3, 55.4, 34.6, 33.6, 33.4 ppm.
MS (ESI): m/z = 391.1 [M + H]+. C22H22N4OS (390.50): calcd. C
67.67, H 5.68, N 14.35, S 8.21; found C 67.76, H 5.53, N 14.44, S
7.92.
General Procedure for the Synthesis of Epiminopyrimido[4,5-b]azep-
ines 3: TsOH (0.025 mmol) was added to a solution of N-allylpyr-
imidinecarbaldehyde 1 (0.5 mmol) and aniline (0.525 mmol) in tol-
uene. The resulting solution was stirred at reflux, and water was
removed using a Dean–Stark trap until 1 had been consumed, as
monitored by TLC. The reaction mixture was concentrated in
vacuo, and the residue was purified by flash column chromatog-
raphy (petroleum ether/EtOAc, 2:1, v/v).
Ethyl 2-[(5R*,8R*)-9-Methyl-4-(phenylthio)-6,7,8,9-tetrahydro-5H-
5,8-epiminopyrimido[4,5-b]azepin-10-yl]acetate (3f): White solid
(0.111 g, 60%). M.p. 62–63 °C. H NMR (300 MHz, CDCl3): δ =
(5R*,8R*)-9-Methyl-10-phenyl-4-(phenylthio)-6,7,8,9-tetrahydro-
5H-5,8-epiminopyrimido[4,5-b]azepine (3a): Pale yellow solid
1
1
(0.175 g, 97%). M.p. 89–90 °C. H NMR (300 MHz, CDCl3): δ =
8.25 (s, 1 H, pyrimidyl-H), 7.55–7.45 (m, 2 H, phenyl-H), 7.43–7.33
8.12 (s, 1 H, pyrimidyl-H), 7.51–7.48 (m, 2 H, phenyl-H), 7.41–7.35
(m, 3 H, phenyl-H), 4.54 (d, J = 4.8 Hz, 1 H, CH), 4.43 (d, J =
(m, 3 H, phenyl-H), 7.24–7.19 (m, 2 H, phenyl-H), 6.90–6.82 (m, 3 6.0 Hz, 1 H, CH), 4.21 (q, J = 7.2 Hz, 2 H, CH2), 3.37 (A of AB,
H, phenyl-H), 5.19 (s, 1 H, CH), 5.18 (s, 1 H, CH), 3.14 (s, 3 H,
J = 16.8 Hz, 1 H, CH2), 3.30 (B of AB, J = 16.8 Hz, 1 H, CH2),
3.07 (s, 3 H, CH3), 2.41–2.30 (m, 2 H, CH2), 2.20–2.05 (m, 1 H,
CH3), 2.43–2.27 (m, 3 H, CH2), 2.14–2.08 (m, 1 H, CH2) ppm. 13
C
NMR (75 MHz, CDCl3): δ = 159.1, 156.8, 156.5, 144.9, 133.9, CH2), 1.98–1.89 (m, 1 H, CH2), 1.28 (t, J = 7.2 Hz, 3 H, CH3)
130.3, 129.3, 129.1, 128.4, 120.4, 117.4, 115.7, 74.5, 55.9, 34.5, 33.6,
33.4 ppm. MS (ESI): m/z = 361.1 [M + H]+. C21H20N4S (360.48):
calcd. C 69.97, H 5.59, N 15.54, S 8.90; found C 70.13, H 5.26, N
15.74, S 8.72.
ppm. 13C NMR (75 MHz, CDCl3): δ = 170.1, 160.0, 156.42, 156.37,
134.2, 129.6, 128.8, 128.4, 112.6, 77.1, 60.7, 58.5, 48.0, 34.3, 33.6,
33.0, 14.0 ppm. MS (ESI): m/z = 371.5 [M + H]+. C19H22N4O2S
(370.47): calcd. C 61.60, H 5.99, N 15.12, S 8.66; found C 61.53,
H 6.02, N 14.87, S 8.40.
(5R*,8R*)-10-(4-Cyanophenyl)-9-methyl-4-(phenylthio)-6,7,8,9-tetra-
hydro-5H-5,8-epiminopyrimido[4,5-b]azepine (3b): White solid
(5R*,8R*)-10-Butyl-9-methyl-4-(phenylthio)-6,7,8,9-tetrahydro-
5H-5,8-epiminopyrimido[4,5-b]azepine (3g): Pale yellow oil (0.104 g,
61%). 1H NMR (300 MHz, CDCl3): δ = 8.23 (s, 1 H, pyrimidyl-
1
(0.164 g, 93%). M.p. 157–158 °C. H NMR (300 MHz, CDCl3): δ
= 8.16 (s, 1 H, pyrimidyl-H), 7.56–7.48 (m, 4 H, phenyl-H), 7.44–
7.39 (m, 3 H, phenyl-H), 6.92 (d, J = 8.7 Hz, 2 H, phenyl-H), 5.23 H), 7.50–7.47 (m, 2 H, phenyl-H), 7.37–7.35 (m, 3 H, phenyl-H),
(s, 1 H, CH), 5.22 (s, 1 H, CH), 3.15 (s, 3 H, CH3), 2.51–2.34 (m, 4.32 (d, J = 6.3 Hz, 1 H, CH), 4.29 (d, J = 4.2 Hz, 1 H, CH), 3.07
3 H, CH2), 2.22–2.15 (m, 1 H, CH2) ppm. 13C NMR (75 MHz, (s, 3 H, CH3), 2.54–2.35 (m, 2 H, CH2), 2.28–2.01 (m, 3 H, CH2),
CDCl3): δ = 159.6, 156.6, 148.7, 134.0, 133.5, 129.5, 129.1, 128.6,
119.2, 117.3, 114.7, 102.7, 74.0, 55.9, 34.2, 33.5, 33.3 ppm. MS
(ESI): m/z = 386.3 [M + H]+. HRMS (ESI-TOF): calcd. for
C22H20N5S [M + H]+ 386.1426; found 386.1434.
1.93–1.86 (m, 1 H, CH2), 1.56–1.46 (m, 2 H, CH2), 1.38–1.30 (m,
2 H, CH2), 0.91 (t, J = 6.6 Hz, 3 H, CH3) ppm. 13C NMR
(75 MHz, CDCl3): δ = 159.1, 156.8, 156.3, 133.9, 130.3, 128.9,
128.2, 113.7, 77.2, 57.1, 45.9, 34.3, 33.4, 33.1, 30.3, 20.5, 13.8 ppm.
MS (ESI): m/z = 341.1 [M + H]+. C19H24N4S (340.49): calcd. C
67.02, H 7.10, N 16.45, S 9.42; found C 67.27, H 6.82, N 16.40, S
9.32.
(5R*,8R*)-9-Methyl-10-(4-nitrophenyl)-4-(phenylthio)-6,7,8,9-
tetrahydro-5H-5,8-epiminopyrimido[4,5-b]azepine (3c): Pale yellow
solid (0.192 g, 95 %). M.p. 200–201 °C. 1H NMR (300 MHz,
CDCl3): δ = 8.15 (s, 1 H, pyrimidyl-H), 8.13 (d, J = 9.0 Hz, 2 H,
(5R*,8R*)-10-Benzyl-9-methyl-4-(phenylthio)-6,7,8,9-tetra-
phenyl-H), 7.52–7.49 (m, 2 H, phenyl-H), 7.41–7.39 (m, 3 H, hydro-5H-5,8-epiminopyrimido[4,5-b]azepine (3h): Pale yellow oil
phenyl-H), 6.92 (d, J = 9.3 Hz, 2 H, phenyl-H), 5.28 (s, 1 H, CH), (0.157 g, 84%). 1H NMR (300 MHz, CDCl3): δ = 8.28 (s, 1 H,
5.26 (s, 1 H, CH), 3.16 (s, 3 H, CH3), 2.53–2.38 (m, 3 H, CH2), pyrimidyl-H), 7.51–7.48 (m, 2 H, phenyl-H), 7.41–7.35 (m, 3 H,
2.24–2.17 (m, 1 H, CH2) ppm. 13C NMR (75 MHz, CDCl3): δ =
phenyl-H), 7.34–7.26 (m, 5 H, phenyl-H), 4.26 (d, J = 5.7 Hz, 1 H,
159.9, 156.8, 156.6, 150.7, 140.6, 134.2, 128.8, 125.73, 125.71, CH), 4.15 (d, J = 4.5 Hz, 1 H, CH), 3.64 (A of AB, J = 12.9 Hz,
116.6, 114.7, 90.5, 74.4, 56.3, 34.3, 33.7, 33.5 ppm. MS (ESI): m/z
= 406.1 [M + H]+. C21H19N5O2S (405.47): calcd. C 62.21, H 4.72,
N 17.27, S 7.91; found C 62.28, H 4.55, N 17.43, S 7.77.
1 H, CH2), 3.59 (B of AB, J = 12.9 Hz, 1 H, CH2), 3.04 (s, 3 H,
CH3), 2.31–2.20 (m, 1 H, CH2), 2.18–2.12 (m, 1 H, CH2), 2.10–
2.01 (m, 1 H, CH2), 1.95–1.88 (m, 1 H, CH2) ppm. 13C NMR
(75 MHz, CDCl3): δ = 159.3, 156.5, 156.1, 137.6, 133.9, 129.9,
128.9, 128.6, 128.0, 126.9, 113.1, 76.1, 57.2, 50.5, 34.1, 33.4,
32.6 ppm. MS (ESI): m/z = 375.1 [M + H]+. HRMS (ESI-TOF):
calcd. for C22H23N4S [M + H]+ 375.1632; found 375.1638.
(5R*,8R*)-9-Methyl-10-(2-nitrophenyl)-4-(phenylthio)-6,7,8,9-
tetrahydro-5H-5,8-epiminopyrimido[4,5-b]azepine (3d): Yellow solid
1
(0.134 g, 66%). M.p. 155–156 °C. H NMR (300 MHz, CDCl3): δ
= 8.19 (s, 1 H, pyrimidyl-H), 7.76 (d, J = 1.5 Hz, 1 H, phenyl-H),
7.55–7.52 (m, 2 H, phenyl-H), 7.41–7.34 (m, 4 H, phenyl-H), 7.02
(5R*,8R*)-9-Allyl-10-phenyl-4-(phenylthio)-6,7,8,9-tetrahydro-5H-
(t, J = 8.1 Hz, 1 H, phenyl-H), 6.90 (d, J = 8.4 Hz, 1 H, phenyl- 5,8-epiminopyrimido[4,5-b]azepine (3i): Pale yellow solid (0.183 g,
1
H), 4.97 (d, J = 5.1 Hz, 1 H, CH), 4.85 (d, J = 4.2 Hz, 1 H, CH),
95%). M.p. 120–121 °C. H NMR (300 MHz, CDCl3): δ = 8.14 (s,
3.04 (s, 3 H, CH3), 2.50–2.44 (m, 2 H, CH2), 2.29–2.24 (m, 1 H,
1 H, pyrimidyl-H), 7.53–7.51 (m, 2 H, phenyl-H), 7.41–7.36 (m, 3
CH2), 2.14–2.09 (m, 1 H, CH2) ppm. 13C NMR (75 MHz, CDCl3): H, phenyl-H), 7.24–7.18 (m, 2 H, phenyl-H), 6.90–6.82 (m, 3 H,
δ = 159.6, 156.7, 156.6, 142.2, 138.3, 134.5, 133.6, 129.1, 129.1,
phenyl-H), 5.78–5.73 (m, 1 H, CH=C), 5.26–5.16 (m, 4 H, 2 CH,
Eur. J. Org. Chem. 2014, 660–669
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
667

Z.-X. Yu, X. Bai et al.
FULL PAPER
C=CH2), 4.29 (ddt, J = 15.3, 6.0, 1.5 Hz, 1 H, CH2), 4.08 (ddt, J
to Z.-X. Y.), a seed research grant from Jilin University, and a grant
= 15.6, 6.6, 1.2 Hz, 1 H, CH2), 2.47–2.38 (m, 2 H, CH2), 2.31–2.25 from the Department of Science and Technology of Jilin Province
(m, 1 H, CH2), 2.15–2.08 (m, 1 H, CH2) ppm. 13C NMR (75 MHz,
CDCl3): δ = 159.6, 156.4, 156.3, 144.5, 133.9, 133.4, 130.2, 129.1,
129.0, 128.4, 120.3, 118.0, 117.5, 115.4, 72.3, 55.3, 48.5, 35.3,
33.3 ppm. MS (ESI): m/z = 386.9 [M + H]+. C23H22N4S (386.51):
calcd. C 71.47, H 5.74, N 14.50, S 8.30; found C 71.37, H 5.60, N
14.58, S 8.15.
of China (No. 20106039). Additional support was provided by
Changchun Discovery Sciences, Ltd. Professor Yazhong Pei is
thanked for helpful discussions during the prepation of this manu-
script.
[1] For reviews, see: a) J. E. Biggs-Houck, A. Younai, J. T. Shaw,
Curr. Opin. Chem. Biol. 2010, 14, 371–382; b) S. Dandapani,
L. A. Marcaurelle, Curr. Opin. Chem. Biol. 2010, 14, 362–370;
c) W. R. J. D. Galloway, A. Isidro-Llobet, D. R. Spring, Nat.
Commun. 2010, 1, 1–13; d) E. Ruijter, R. Scheffelaar, R. V. A.
Orru, Angew. Chem. 2011, 123, 6358–6371; Angew. Chem. Int.
Ed. 2011, 50, 6234–6246; e) C. J. O’Connor, H. S. G.
Beckmann, D. R. Spring, Chem. Soc. Rev. 2012, 41, 4444–4456.
[2] a) E. M. Beccalli, G. Broggini, G. Paladino, A. Penoni, C. Zoni,
J. Org. Chem. 2004, 69, 5627–5630; b) G. Cuny, M. Bois-
Choussy, J. Zhu, J. Am. Chem. Soc. 2004, 126, 14475–14484;
c) T. Xiong, Q. Zhang, Z. Zhang, Q. Liu, J. Org. Chem. 2007,
72, 8005–8009; d) V. A. Chebanov, V. E. Saraev, S. M. Desenko,
V. N. Chernenko, I. V. Knyazeva, U. Groth, T. N. Glasnov,
C. O. Kappe, J. Org. Chem. 2008, 73, 5110–5118; e) D. S. Erm-
olat’ev, E. V. Van der Eycken, J. Org. Chem. 2008, 73, 6691–
6697; f) S. Kumar, H. Ila, H. Junjappa, J. Org. Chem. 2009,
74, 7046–7051; g) S. F. Vasilevsky, T. y. F. Mikhailovskaya, V. I.
Mamatyuk, G. E. Salnikov, G. A. Bogdanchikov, M. Mano-
haran, I. V. Alabugin, J. Org. Chem. 2009, 74, 8106–8117; h)
S.-L. Wang, Y.-P. Liu, B.-H. Xu, X.-H. Wang, B. Jiang, S.-J.
Tu, Tetrahedron 2011, 67, 9417–9425.
(5R*,8R*)-4-Chloro-9-methyl-10-phenyl-6,7,8,9-tetrahydro-5H-
5,8-epiminopyrimido[4,5-b]azepine (3j): White solid (0.116 g, 80%).
1
M.p. 174–176 °C. H NMR (300 MHz, CDCl3): δ = 8.11 (s, 1 H,
pyrimidyl-H), 7.22–7.19 (m, 2 H, phenyl-H), 6.88–6.83 (m, 3 H,
phenyl-H), 5.21 (d, J = 5.1 Hz, 1 H, CH), 5.13 (d, J = 6.3 Hz, 1
H, CH), 3.16 (s, 3 H, CH3), 2.51–2.39 (m, 2 H, CH2), 2.32–2.29
(m, 1 H, CH2), 2.20–2.21 (m, 1 H, CH2) ppm. 13C NMR (75 MHz,
CDCl3): δ = 158.6, 156.4, 147.6, 144.4, 129.4, 120.7, 117.2, 114.9,
74.8, 56.3, 34.5, 33.7, 33.4 ppm. MS (ESI): m/z = 287.0 [M + H]+.
C15H15ClN4 (286.76): calcd. C 62.83, H 5.27, N 19.54; found C
62.98, H 5.30, N 19.50.
(5R*,8R*)-9-Methyl-4-phenoxy-10-phenyl-6,7,8,9-tetrahydro-5H-
5,8-epiminopyrimido[4,5-b]azepine (3k): Yellow solid (0.115 g,
1
67%). M.p. 67–69 °C. H NMR (300 MHz, CDCl3): δ = 8.06 (s, 1
H, pyrimidyl-H), 7.43–7.38 (m, 2 H, phenyl-H), 7.24–7.19 (m, 3 H,
phenyl-H), 7.11–7.08 (m, 2 H, phenyl-H), 6.94–6.84 (m, 3 H,
phenyl-H), 5.25 (d, J = 6.0 Hz, 1 H, CH), 5.22 (d, J = 4.5 Hz, 1
H, CH), 3.21 (s, 3 H, CH3), 2.53–2.33 (m, 3 H, CH2), 2.28–2.22
(m, 1 H, CH2) ppm. 13C NMR (75 MHz, CDCl3): δ = 156.1, 153.7,
129.5, 129.3, 127.3, 124.9, 124.8, 123.2, 121.2, 121.0, 120.4, 117.4,
74.5, 53.4, 34.5, 34.3, 33.9 ppm. MS (ESI): m/z = 345.0 [M + H]+.
HRMS (ESI-TOF): calcd. for C21H21N4O [M + H]+ 345.1702;
found 345.1710.
[3] For reviews, see: a) K. A. Jørgensen, Angew. Chem. 2000, 112,
3702–3733; Angew. Chem. Int. Ed. 2000, 39, 3558–3588; b) P.
Buonora, J.-C. Olsen, T. Oh, Tetrahedron 2001, 57, 6099–6138;
c) V. V. Kouznetsov, Tetrahedron 2009, 65, 2721–2750.
[4] For a review, see: R. M. Borzilleri, S. M. Weinreb, Synthesis
1995, 347–360.
(5R*,8R*)-4-Methoxy-9-methyl-10-phenyl-6,7,8,9-tetrahydro-5H-
5,8-epiminopyrimido[4,5-b]azepine (3l): Yellow solid (0.099 g, 70%).
[5] For reviews, see: a) H. M. R. Hoffmann, Angew. Chem. 1969,
81, 597–618; Angew. Chem. Int. Ed. Engl. 1969, 8, 556–577; b)
K. Mikami, M. Shimizu, Chem. Rev. 1992, 92, 1021–1050.
[6] a) B. B. Snider, Acc. Chem. Res. 1980, 13, 426–432; b) M. Yam-
anaka, A. Nishida, M. Nakagawa, Org. Lett. 2000, 2, 159–161;
c) M. Yamanaka, A. Nishida, M. Nakagawa, J. Org. Chem.
2003, 68, 3112–3120; d) M. Terada, K. Machioka, K. Sorima-
chi, Angew. Chem. 2006, 118, 2312–2315; Angew. Chem. Int.
Ed. 2006, 45, 2254–2257; e) M. L. Clarke, M. B. France, Tetra-
hedron 2008, 64, 9003–9031; f) L. H. Oliver, L. A. Puls, S. L.
Tobey, Tetrahedron Lett. 2008, 49, 4636–4639; g) M. Rueping,
T. Theissmann, A. Kuenkel, R. M. Koenigs, Angew. Chem.
2008, 120, 6903–6906; Angew. Chem. Int. Ed. 2008, 47, 6798–
6801; h) B. Han, Y.-C. Xiao, Y. Yao, Y.-C. Chen, Angew. Chem.
2010, 122, 10387–10389; Angew. Chem. Int. Ed. 2010, 49,
10189–10191; i) Y.-J. Zhao, B. Li, L.-J. S. Tan, Z.-L. Shen, T.-
P. Loh, J. Am. Chem. Soc. 2010, 132, 10242–10244; j) G. Hilt,
F. Erver, K. Harms, Org. Lett. 2011, 13, 304–307.
[7] a) T. Inazumi, E. Harada, T. Mizukoshi, Y. Kuroki, A. Kakehi,
M. Noguchi, J. Chem. Soc. Perkin Trans. 1 1994, 565–570; b)
T. Inazumi, K. Yamada, Y. Kuroki, A. Kakehi, M. Noguchi,
J. Chem. Soc. Perkin Trans. 1 1994, 557–564; c) Y. Kuroki, R.
Akao, T. Inazumi, M. Noguchi, Tetrahedron 1994, 50, 1063–
1072; d) M. Noguchi, T. Mizukoshi, A. Kakehi, Tetrahedron
1996, 52, 13081–13096; e) M. Noguchi, T. Mizukoshi, S. Naka-
gawa, A. Kakehi, Tetrahedron 1996, 52, 13111–13120; f) M.
Noguchi, T. Mizukoshi, T. Uchida, Y. Kuroki, Tetrahedron
1996, 52, 13097–13110; g) M. Noguchi, R. Matsushita, S. Tak-
amura, T. Uchida, A. Kakehi, M. Shiro, H. Yamamoto, Tetra-
hedron Lett. 2000, 41, 8489–8493; h) M. Noguchi, H. Yamada,
S. Takamura, K. Okada, A. Kakehi, H. Yamamoto, Tetrahe-
dron 2000, 56, 1299–1307; i) M. Noguchi, H. Yamada, S. Taka-
mura, T. Uchida, M. Hironaka, A. Kakehi, H. Yamamoto, Eur.
J. Org. Chem. 2000, 1489–1496; j) S. Takamura, H. Yamada,
1
M.p. 119–120 °C. H NMR (300 MHz, CDCl3): δ = 8.09 (s, 1 H,
pyrimidyl-H), 7.19–7.15 (m, 2 H, phenyl-H), 6.86–6.78 (m, 3 H,
phenyl-H), 5.15 (d, J = 4.2 Hz, 1 H, CH), 5.02 (d, J = 5.4 Hz, 1
H, CH), 3.94 (s, 3 H, OCH3), 3.13 (s, 3 H, CH3), 2.42–2.22 (m, 3
H, CH2), 2.15–2.08 (m, 1 H, CH2) ppm. 13C NMR (75 MHz,
CDCl3): δ = 164.0, 158.5, 155.6, 145.3, 129.1, 120.0, 117.3, 100.1,
74.3, 53.4, 53.0, 34.6, 34.2, 33.6 ppm. MS (ESI): m/z = 283.0 [M +
H]+. C16H18N4O (282.34): calcd. C 68.06, H 6.43, N 19.84; found
C 68.01, H 6.36, N 19.92.
(5R*,8R*)-N,N-Diethyl-9-methyl-10-phenyl-6,7,8,9-tetrahydro-5H-
5,8-epiminopyrimido[4,5-b]azepin-4-amine (3m): Pale yellow solid
1
(0.071 g, 44%). M.p. 74–75 °C. H NMR (300 MHz, CDCl3): δ =
8.09 (s, 1 H, pyrimidyl-H), 7.13–7.10 (m, 2 H, phenyl-H), 6.80–6.75
(m, 3 H, phenyl-H), 5.15 (d, J = 4.2 Hz, 1 H, CH), 4.88 (d, J =
6.6 Hz, 1 H, CH), 3.57–3.46 (m, 2 H, CH2), 3.13–3.10 (m, 5 H,
CH3, CH2), 2.58–2.50 (m, 1 H, CH2), 2.39–2.32 (m, 2 H, CH2),
2.26–2.18 (m, 1 H, CH2), 1.18 (t, J = 6.9 Hz, 6 H, 2 CH3) ppm.
13C NMR (75 MHz, CDCl3): δ = 161.9, 158.0, 155.2, 145.1, 128.9,
124.4, 120.0, 117.4, 104.6, 74.5, 54.9, 45.3, 34.3, 34.0, 13.7 ppm.
MS (ESI): m/z = 324.0 [M + H]+. C19H25N5 (323.44): calcd. C
70.56, H 7.79, N 21.65; found C 70.80, H 7.76, N 21.88.
Supporting Information (see footnote on the first page of this arti-
cle): Copies of LC–mass and NMR spectra for all products, and
additional computational results and discussion.
Acknowledgments
This work was supported by grants from the National Natural Sci-
ence Foundation of China (No. 81072526 to X. B., and 21072013
668
www.eurjoc.org
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2014, 660–669

Switching between Imino Diels–Alder and Ene-Type Cyclization
T. Michinaka, H. Yamamoto, A. Kakehi, M. Noguchi, J.
Chem. Res. Synop. 2000, 426–428; k) M. Noguchi, T. Suna-
gawa, R. Akao, H. Yamada, H. Yamamoto, A. Kakehi, Tetra-
hedron 2007, 63, 4548–4557.
v. R. Schleyer, J. A. Pople (Eds.), Ab Initio Molecular Orbital
Theory, Wiley, New York, 1986.
[13] a) V. Barone, M. Cossi, J. Phys. Chem. A 1998, 102, 1995–
2001; b) M. Cossi, N. Rega, G. Scalmani, V. Barone, J. Comput.
Chem. 2003, 24, 669–681; c) Y. Takano, K. N. Houk, J. Chem.
Theory Comput. 2005, 1, 70–77.
[8] V. Gaddam, R. Meesala, R. Nagarajan, Synthesis 2007, 2503–
2512.
[9] a) J. Yang, X. Che, Q. Dang, Z. Wei, S. Gao, X. Bai, Org. Lett.
2005, 7, 1541–1543; b) J. Liu, Q. Dang, Z. Wei, F. Shi, X. Bai,
J. Comb. Chem. 2006, 8, 410–416; c) R. Fu, X. Xu, Q. Dang,
F. Chen, X. Bai, Org. Lett. 2007, 9, 571–574; d) X. Che, L.
Zheng, Q. Dang, X. Bai, J. Org. Chem. 2008, 73, 1147–1149;
e) J. Xiang, H. Xie, D. Wen, Q. Dang, X. Bai, J. Org. Chem.
[14] a) K. Fukui, J. Phys. Chem. 1970, 74, 4161–4163; b) C. Gonza-
lez, H. B. Schlegel, J. Chem. Phys. 1989, 90, 2154–2161; c) C.
Gonzalez, H. B. Schlegel, J. Phys. Chem. 1990, 94, 5523–5527.
[15] In a computational study of a similar ene-type reaction, an
intermediate formed from the 1,6-H-shift (ene-type) reaction
was located.[7f]
2008, 73, 3281–3283; f) L. Y. Zheng, F. Z. Yang, Q. Dang, X. [16] We also considered the diradical character of IN3. However,
Bai, Org. Lett. 2008, 10, 889–892; g) J. Xiang, D. Wen, H. Xie,
Q. Dang, X. Bai, J. Comb. Chem. 2010, 12, 503–509.
[10] F. Yang, L. Zheng, J. Xiang, Q. Dang, X. Bai, J. Comb. Chem.
2010, 12, 476–481.
we could only locate the triplet state of IN3, namely IN3Ј [opti-
mized by using UB3LYP/6–311+G(d,p)], but not its singlet di-
radical species. Therefore, we can rule out this diradical path-
way. Another point that rules out the possible diradical path-
way is that the single-point energy of the triplet state of IN3Ј
is higher than the energy of TS2 by 11.0 kcal/mol. This indi-
cates that if a diradical intermediate like IN3Ј could exist, it
should be higher in energy than TS2 and would become disfa-
vored. The possible diradical IADA from 4a can also be ex-
cluded by the same arguments. Details can be found in the
Supporting Information.
[11] M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria,
M. A. Robb, J. R. Cheeseman, J. A. Montgomery Jr., T.
Vreven, K. N. Kudin, J. C. Burant, J. M. Millam, S. S. Iyengar,
J. Tomasi, V. Barone, B. Mennucci, M. Cossi, G. Scalmani, N.
Rega, G. A. Petersson, H. Nakatsuji, M. Hada, M. Ehara, K.
Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y.
Honda, O. Kitao, H. Nakai, M. Klene, X. Li, J. E. Knox, H. P.
Hratchian, J. B. Cross, C. Adamo, J. Jaramillo, R. Gomperts,
R. E. Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pom-
[17] X. Li, S. Ye, C. He, Z.-X. Yu, Eur. J. Org. Chem. 2008, 4296–
4303.
elli, J. W. Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. [18] a) For a DFT study on the [4+ + 2] reactions between N-pro-
Salvador, J. J. Dannenberg, V. G. Zakrzewski, S. Dapprich,
A. D. Daniels, M. C. Strain, O. Farkas, D. K. Malick, A. D.
Rabuck, K. Raghavachari, J. B. Foresman, J. V. Ortiz, Q. Cui,
A. G. Baboul, S. Clifford, J. Cioslowski, B. B. Stefanov, G. Liu,
A. Liashenko, P. Piskorz, I. Komaromi, R. L. Martin, D. J.
Fox, T. Keith, M. A. Al-Laham, C. Y. Peng, A. Nanayakkara,
M. Challacombe, P. M. W. Gill, B. Johnson, W. Chen, M. W.
Wong, C. Gonzalez, J. A. Pople, Gaussian 03, Revision C.02,
Gaussian, Inc., Wallingford, CT, 2004.
tonated azadienes and olefins, see: Y.-Q. Ding, D.-C. Fang, J.
Org. Chem. 2003, 68, 4382–4387; b) Y. Liang, X. Jiang, Z.-X.
Yu, Org. Lett. 2009, 11, 5302–5305; c) for DFT studies on [4
+ 2+] reactions of iminium cations, see: d) L. R. Domingo, J.
Org. Chem. 2001, 66, 3211–3214; e) R. G. Iafe, K. N. Houk, J.
Org. Chem. 2008, 73, 2679–2686; f) L. R. Domingo, J. A. Saez,
Org. Biomol. Chem. 2009, 7, 3576–3583.
[19] We can exclude the diradical pathways of the IADA reactions
in Figure 1b. The diradical intermediate from 5a is higher in
energy than exo-HTS1 by 15.9 kcal/mol.
[12] For B3LYP: a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648–
5652; b) C. Lee, W. Yang, R. G. Parr, Phys. Rev. B 1988, 37,
785–789; c) for 6–311+G(d,p): W. J. Hehre, L. Radom, P.
Received: August 30, 2013
Published Online: November 19, 2013
Eur. J. Org. Chem. 2014, 660–669
© 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
669