Pd-catalyzed cycloisomerization of 4-aza-1,6-enynes to 3-aza-bicyclo[4.1.0] hept-2-enes

Article abstract of DOI:10.1002/chem.201402087

A method for the synthesis of bicyclo[4.1.0]heptenes from 1,6-enynes through Pd-catalyzed cycloisomerization has been developed. N- and O-tethered 1,6-enynes were successfully transformed to their corresponding 3-aza- and 3-oxabicyclo[4.1.0]heptenes in reasonable-to-high yields using the catalysts [PdCl₂(CH₃CN)₂]/P(OPh)₃ or [Pd(maleimidate)₂(PPh₃)₂] in toluene. The computational calculations using density functional theory indicate that [PdCl₂{P(OPh)₃}] in the oxidation state Pd^II acts as the active catalyst species for the formation of 3-azabicyclo[4.1.0] heptenes through 6-endo-dig cyclization. Pd-catalyzed cycloisomerization of 1,6-enynes has long been studied and provides a variety of 5-membered ring systems. A method that transforms 1,6-enynes to bicyclo[4.1.0]heptene-type products through palladium catalysis was elusive but eventually has been found. A broad range of 4-aza-1,6-enynes were successfully transformed to 3-aza-bicyclo[4.1.0]heptenes in moderate-to-high yields when [PdCl ₂]/P(OPh)₃ or [Pd(maleimidate)₂(PPh ₃)₂] catalysts were used in a nonpolar solvent (see figure).

Full text of DOI:10.1002/chem.201402087

DOI: 10.1002/chem.201402087
Full Paper
&
Palladium
Pd-Catalyzed Cycloisomerization of 4-Aza-1,6-enynes to
3-Aza-bicyclo[4.1.0]hept-2-enes
Hye Mi Oh,[a] Ji Eun Park,[b] Jisu Kim,[a] Ju Hyun Kim,[a] Youn K. Kang,*[b] and
Young Keun Chung*[a]
Abstract:
A
method for the synthesis of bicyclo-
[Pd(maleimidate)2(PPh3)2] in toluene. The computational cal-
culations using density functional theory indicate that
[PdCl2{P(OPh)3}] in the oxidation state PdII acts as the active
catalyst species for the formation of 3-azabicyclo-
[4.1.0]heptenes through 6-endo-dig cyclization.
[4.1.0]heptenes from 1,6-enynes through Pd-catalyzed cycloi-
somerization has been developed. N- and O-tethered 1,6-
enynes were successfully transformed to their corresponding
3-aza- and 3-oxabicyclo[4.1.0]heptenes in reasonable-to-high
yields using the catalysts [PdCl2(CH3CN)2]/P(OPh)3 or
Introduction
Transition-metal-catalyzed cycloisomerization reactions have
attracted a great deal of attention because they can provide
a variety of polyfunctional cyclic compounds in a simple and
easy manner.[1] In particular, the cycloisomerization of 1,6-
enynes has been proved to afford a wide variety of cyclic com-
pounds.[2] Although several transition-metal complexes pro-
mote the cycloisomerization of 1,6-enynes, an efficient catalytic
system still remains an ongoing challenge.
Pd-catalyzed reactions are widely used in modern organic
chemistry.[3] Their applicability has been expanded to the cyclo-
isomerization of 1,n-enynes since the pioneering work by Laut-
ens and Trost in 1985,[4a] and significant developments have
been achieved in this field.[4,5] Many different types of Pd cata-
lysts have been applied to 1,6-enynes, leading to the formation
of a variety of products as follows: Alder-ene-type products
such as dialkylidene cyclopentanes (II) and alkenyl alkylidene
cyclopentanes (III), enyne metathesis products from vinylcyclo-
pentenes (IV), hydroxy- or alkoxycyclization products (V),
Scheme 1. Pd-catalyzed cycloisomerization reactions of 1,6-enyne.
bicyclo[3.2.0]heptenes
(VI),
and
cyclopentenylbicyclo-
[3.1.0]hexanes (VII), as shown in Scheme 1.
Yamamoto reported that the combination of a Pd0 precata-
lyst and triphenyl phosphite could afford new types of cycloi-
somerization products other than the normal five-membered
products, as described above.[6a] Recently, Liang has reported
that vinylidenepyridines or vinylidenepyrrolidines can be pre-
pared from 1,6-enyne carbonates using Pd catalysts.[6b] These
studies clearly demonstrate that the Pd-catalyzed cycloisomeri-
zation of 1,n-enynes (n=6 and 7) is an efficient reaction for
the synthesis of five- and six-membered carbo- and heterocy-
cles.
[a] H. M. Oh,+ J. Kim, J. H. Kim, Prof. Y. K. Chung
Department of Chemistry, College of Natural Sciences
Seoul National University, Seoul 151-747 (Korea)
E-mail: ykchung@snu.ac.kr
Despite all these efforts, however, the transformation of 1,6-
enynes to bicyclo[4.1.0]heptene-type product (I, Scheme 1) in
the presence of palladium catalyst has not been reported yet
to the best of our knowledge, although other transition metals
such as Pt (20–97%),[7] Au (66–78%),[8] Ir (15–84%),[9] and Rh
(20–95%)[10] have been recognized to perform such a transfor-
mation.
[b] J. E. Park,+ Prof. Y. K. Kang
Department of Chemistry
Sangmyung University, Seoul 110-743 (Korea)
E-mail: younkang@smu.ac.kr
[+] These authors contributed equally to this work.
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201402087.
Chem. Eur. J. 2014, 20, 1 – 14
1
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Recently, we found that the cycloisomerization of 4-aza-1,6-
enynes in the presence of the Pd0/maleimide or PdII/P(OPh)3
catalysts, afforded 3-aza-bicyclo[4.1.0]heptenes; this type of
transformation has not been reported. Given our ongoing in-
terest in metal-catalyzed cyclizations of 1,6-enynes[11] and the
synthesis of bicyclo[4.1.0]heptenes, particularly for their poten-
tial to afford new compounds,[12] this unprecedented result
was quite intriguing and encouraged us to explore further.
Herein, we report the experimental results as well as the mech-
anistic rationale from computational calculations.
mide did not produce any cycloisomerized product (Table 1,
entry 6). Similar results were obtained when N-phenylmalei-
mide was used in which the nitrogen is protected, indicating
that imidate formation was not possible (Table 1, entry 7).
Moreover, a series of dione compounds that are known to
form Pd catalysts in the Pd0 oxidation state afforded the corre-
sponding products (1b) in low yields (Table 1, entries 8–11).
The results showed that maleimide acted as a ligand in the
imidate form to produce a PdII species, probably [Pd(mal)2-
(PPh3)2] (mal=maleimidate, C4H2NO2).
To confirm this hypothesis, [Pd(mal)2(PPh3)2] was prepared
according to a literature procedure[14] and used as the catalyst
in the reaction. A test reaction of 1a using this separately pre-
pared catalyst under the same reaction conditions as in entry 4
in Table 1 afforded 1c in 62% yield (data not shown in
Table 1), which is slightly higher than that produced using the
[Pd(PPh3)4]/maleimide catalyst system (Table 1, entry 4, 51%
yield), indicating that [Pd(mal)2(PPh3)2] is the precursor of the
active catalytic species. With this confirmed catalyst in hand,
the reaction solvent and temperature were optimized
Results and Discussion
Optimization of the reaction conditions
Our initial study began with the reaction of enyne (1a) in the
presence of [Pd(PPh3)4] (10 mol%) in 1,4-dioxane at 908C;
a monocyclic triene compound (1b) was isolated in 21% yield
[Eq. (1)].
(Table 1, entries 12–16). The reaction was found to be
highly sensitive to the type of solvent used, and tolu-
ene gave the best result at 1308C. At this tempera-
ture, a reaction time of 7 h was enough to obtain the
maximum yield of 1c.
In spite of the optimized reaction conditions, the
catalyst system showed limited efficiency, for exam-
To use the in situ-generated triene 1b for the consecutive
one-pot Diels–Alder reaction, maleimide (1.3 equiv) was added
at the beginning of the reaction. However, the expected Diels–
Alder product was not observed. Instead, 3-aza-bicyclo-
[4.1.0]heptene (1c) was isolated in 40% yield. This result was
unprecedented; therefore, the reaction conditions were opti-
mized for the synthesis of 1c.
First, the catalyst precursor was changed from [Pd(PPh3)4] to
[Pd2(dba)3] (dba=dibenzylideneacetone); however, the expect-
ed cycloisomerized product was
ple, when (E)-N-(3-cyclopropylallyl)-4-methyl-N-(pent-2-yn-1-yl)-
benzene sulfonamide (2a) was used as the substrate, com-
pound 2c was isolated in 17% yield [Eq. (2)]. Assuming
[PdX2L2] to be the precursor of the active catalytic species, we
have searched for a more general catalyst system. First, [PdCl2-
(MeCN)2] was examined, because it is used in a wide variety of
catalytic reactions.[15] However, in the absence of any additive
(Table 2, entry 2), the reaction did not proceed. When [PdCl2-
(MeCN)2] was used in the presence of phosphine or phosphite,
not obtained (Table 1, entry 3).
Table 1. Palladium-catalyzed cycloisomerization of 1,6-enynes 1a.
When the solvent was changed
from polar 1,4-dioxane to non-
polar toluene, compound 1c
was obtained in 51% yield
Entry Pd complex
([mol%])
Ligand
([mol%])
Solvent
T
[8C]
t
[h]
Yield
1b [%] 1c [%]
Yield
1
[Pd(PPh3)4] (10)
–
dioxane 90
dioxane 90
dioxane 90
toluene 90
toluene 90
toluene 90
toluene 90
toluene 90
toluene 90
23
48[a]
24
15
20
24
24
24
24
16
5
21
0
0
0
0
0
40
0
51
44
0
0
0
0
0
0
9
6
13
84
(Table 1,
entry 4).
At
this
2
3
4
5
[Pd(PPh3)4] (10)
[Pd2(dba)3] (5)
[Pd(PPh3)4] (10)
[Pd(PPh3)4] (10)
maleimide (130)
maleimide (130)
maleimide (40)
succinimide (40)
imidazole (40)
moment, the Pd precursor-to-ad-
ditive ratio was optimized, and
the optimum [Pd(PPh3)4]/malei-
mide ratio was found to be 1:4
for this catalyst system.[13] To de-
termine the nature of the active
catalytic species, the additives
were varied in the reaction.
When maleimide was changed
to succinimide, compound 1c
was isolated in 44% yield
(Table 1, entry 5). However, imi-
dazole with a much higher pKa
value than maleimide or succini-
6
[Pd(PPh3)4] (10)
0
0
7
8
9
[Pd(PPh3)4] (10)
[Pd(PPh3)4] (10)
[Pd(PPh3)4] (10)
N-phenylmaleimide (40)
benzoquinone (40)
furan-2,5-dione (40)
10
29
16
37
0
82
13
0
10
11
12[b]
13[b]
14[b]
15[b]
[Pd(PPh3)4] (10)
[Pd(PPh3)4] (10)
[Pd(C4H2NO2)2(PPh3)2] (10)
[Pd(C4H2NO2)2(PPh3)2] (10)
[Pd(C4H2NO2)2(PPh3)2] (10)
[Pd(C4H2NO2)2(PPh3)2] (10)
cyclopent-4-ene-1,3-dione (40) toluene 90
1,2-cyclohexanedione (40)
–
–
–
–
toluene 90
THF
DCE
dioxane 90
toluene 130[c]
50
70
7
7
7
7
[a] When the reaction time was 24 h, the yield of b and c were 0 and 9%, respectively. [b] [Pd(C4H2NO2)2(PPh3)2]
was prepared according to the literature.[14] [c] When the reaction temperature was 1108C, the yield of b and
c were 0 and 62%, respectively.
&
&
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
2
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Scope of the Pd-catalyzed cycloisomerization for
the synthesis of bicyclo[4.1.0]heptene-type
products
Table 2. Palladium-catalyzed cycloisomerization of 1,6-enynes 2a.
Entry Pd complex
([mol%])
Ligand
([mol%])
Solvent
T
t
Yield Yield
[8C] [h] 1b 1c
[%] [%]
With the optimized reaction conditions in hand, the
substrate scope was examined (Table 3). Most of the
substrates afforded 3-aza-bicyclo[4.1.0]heptenes as
a single diastereomer in reasonable to high yields. An
Alder-ene-type reaction was observed in some cases
(Table 3, entries 3, 7, and 8). The yields of 3-aza-
bicyclo[4.1.0]heptenes were highly dependent on the
substituent on the alkyne and alkene moieties.
1[a]
[Pd(C4H2NO2)2(PPh3)2]
(10)
[PdCl2(MeCN)2] (10)
[PdCl2(MeCN)2(10)
PdCl2(MeCN)2] (10)
[PdCl2(MeCN)2] (10)
–
–
toluene 130 7 18
17
2
3
4
5
toluene 130 24
0
4
0
11
9
triphenylphosphine (40) toluene 130 24
tris-2-furylphosphine (40) toluene 130 24 15
tricyclohexylphosphine
(40)
tris(4-fluorophenyl)phos- toluene 130 24 22
phine (40)
toluene 130 24
5
30
6
[PdCl2(MeCN)2] (10)
17
Enynes with
a terminal alkyne moiety (Table 3,
7
8
9
10
11
[PdCl2(MeCN)2] (10)
[PdCl2(MeCN)2] (10)
[PdCl2(MeCN)2] (10)
[PdCl2(MeCN)2] (10)
[PdCl2(MeCN)2] (10)
tri-o-tolylphoshpine (40) toluene 130 24 18
15
56
0
0
30
triphenylphosphite (40)
triphenylphosphite (40)
triphenylphosphite (40)
triphenylphosphite (40)
toluene 130 24
DCE 70 24
dioxane 90 24
xylene 150
0
0
0
0
entry 3) or sterically bulky substituent on the alkyne
(Table 3, entry 12) gave poor yields and thus may be
considered as the inert substrates under these reac-
tion conditions. When enynes with a cyclopropyl
group at the terminal position of the olefin group
were used as the substrates (Table 3, entries 1–6), the
yield was found to be sensitive to the substituent at
7
[a] [Pd(C4H2NO2)2(PPh3)2] was prepared according to the literature.[14]
a mixture of 2b and 2c in various ratios was isolated (Table 2,
entries 3–8), probably owing to the catalytic activity of
[PdCl2L2] (L=phosphine or phosphite). Among the phosphines
and phosphites used, P(OPh)3 gave the best result. Being too
hydrophilic, [PdCl2{P(OPh)3}2] could not be isolated under ambi-
ent conditions. Thus, [PdCl2(MeCN)2] with four equivalents of
P(OPh)3 (1:4) was used as the catalyst system.
the terminal position of the alkyne in the following order: H !
cyclohexenylꢀEt<CH2OMe<Me<cyclopropyl. In the case of
enynes bearing a 4-MeOC6H4 group at the terminal position of
the olefin (Table 3, entries 9–15), the yield was highly sensitive
to the substituent at the terminal position of the alkyne in the
following order: cyclobutyl<C6H4ꢁOMe-4<n-PrꢀMe<cyclo-
propyl<Et. Substrates 4a and 8a with a methyl and cyclo-
propyl group at the different terminal positions of the alkyne
and olefin gave significantly different yields, that is, 72 and
53% yields, respectively. The enynes with an aryl group at the
internal carbon of the olefin moiety (Table 3, entries 20 and 22)
were better substrates than those (Table 3, entries 18 and 9)
bearing an aryl group at the terminal position of the olefin. In
the cases of enynes with an aryl group at the internal carbon
of the olefin moiety (Table 3, entry 20 vs. 21; 22 vs. 23), the
electronic effect of the substituent on the aromatic ring affect-
ed the yield, for example, the presence of an electron-with-
drawing group diminished the yield of the reaction.
The effect of solvent on the yield of 2c was next examined
Table 2, entries 8–11). The yield of 2c was found to be highly
sensitive to the type of solvent, and toluene gave the best
result (Table 2, entry 8). The reaction did not proceed in 1,2-di-
chloroethene (DCE) or 1,4-dioxane (Table 2, entries 9 and 10). A
higher reaction temperature using xylene (1508C) caused the
decomposition of the product (Table 2, entry 11). Further, a di-
phosphine such as 2,2’-bis(diphenylphosphino)-1,1’-binaphthyl
(BINAP) was examined as an additional ligand in the presence
of [PdCl2(MeCN)2]. However, the reaction produced numerous
unidentified side-products and the major product could not be
isolated (data not shown). Thus, the optimal reaction condi-
tions were determined to be [PdCl2(MeCN)2]/P(OPh)3 (1:4) as
the catalyst, toluene as the solvent, and a temperature of
1308C. When 1a was reacted under these optimized reaction
conditions, compound 1c was obtained as the only product in
62% yield, which is slightly lower than that obtained using
[Pd(mal)2(PPh3)2] yet still considered as an efficient reaction.
Interestingly, the compounds with a cyclopenta[c]pyridine
skeleton were obtained in 41 and 31% yields (Table 3, en-
tries 16 and 19), respectively [Eq. (3)]. A similar type of transfor-
mation was previously observed[11a] in the successive reactions
of a [PtCl2]-catalyzed cycloisomerization of dienynes with
a phenyl group at the terminal carbon of the alkene moiety,
followed by the vinylcyclopropane rearrangement.
Unfortunately, bicyclo[4.1.0]heptenes could not be obtained
from O-tethered enynes using [PdCl2(MeCN)2]/P(OPh)3. Howev-
er, [Pd(mal)2(PPh3)2] could transform O-tethered enynes to their
corresponding 3-oxo-bicyclo[4.1.0]heptenes in relatively low
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
3
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Table 3. Scope of Pd-catalyzed cycloisomerization.[a]
Entry
Substrate
t
[h]
Product
yield [%][b]
Entry
Substrate
t
[h]
Product
yield [%][b]
1
2
1a (R=CH2OMe)
2a (R=Et)
24
24
1c (62)
2c (56)
16a
24
16d (31, R=MeOC6H4)[h]
16
17
18
17a (R=Me)
18a (R=cyclopropyl)
24
24
17c (53)
18c (45)
3[c]
3a (R=H)
19
24
24
3c (4)
4
4a (R=Me)
4c (72)
5c (84)
5[d]
5a (R=cyclopropyl)
19
19a
24
19d (41)
6
7[c]
6a
24
24
6c (53)[e]
7c (10)
20[i]
21[i]
20a (R=C6H5)
21a (R=FC6H4)
12
12
20c (83)
21c (79)
7a (R=Me)
8[c]
8a (R=cyclopropyl)
24
8c (53)
22
23
22a (R=MeOC6H4)
23a (R=CF3C6H4)
12
12
22c (79)
23c (50)
9
9a (R=Me)
10a (R=Et)
24
24
9c (72)
10
10c (89)
11
12
13
14
15
11 a (R=nPr)
12a (R=tBu)
13a (R=cyclopropyl)
14a (R=cyclobutyl)
15a (R=MeOC6H4)
24
24
24
24
24
11 c (71)
12c (0)[f]
13c (77)
14c (56)
15c (63)[g]
24[j]
25[j]
24a (R=Et)
25a (R=nPr)
48
48
24c (37)
21c (20)
[a] Reaction conditions: [PdCl2(MeCN)2] (10 mol%)/P(OPh)3 (40 mol%), toluene, 1308C, 24 h. [b] Yield of the isolated product. [c] Compound 3b, 7b, and
8b were isolated in 16, 8, and 8% yields, respectively. [d] [PdCl2(MeCN)2] (15 mol%) was used. [e] Compound 6a was recovered in 30%. [f] No reaction was
observed. [g] Reactions were carried out at 908C [h] Compound 16a was recovered in 18%. [i] Reactions were carried out at 808C. [j] Compounds 24a and
25a were recovered in 36 and 11% yield, respectively.
yields (Table 3, entries 24 and 25: 37 and 20% yields, respec-
tively). A similar observation has been reported in the Ir-cata-
lyzed cycloisomerization of 1,6-enynes.[9] In the case of C-teth-
ered enynes, both the [PdCl2(MeCN)2]/P(OPh)3 and [Pd(mal)2-
(PPh3)2] catalyst systems did not afford bicyclo[4.1.0]heptenes.
Thus, the bicyclo[4.1.0]heptenes could only be obtained from
heteroatom-tethered enynes, similar to other transition-metal-
catalyzed reactions.[16] The structure of the representative 3-
aza-bicyclo[4.1.0]heptenes, compounds 2c[17] (Figure 1) and
15c[17] (not shown) were confirmed by X-ray crystallographic
analyses.
Mechanistic investigation through computational
calculations
The reaction mechanism was investigated by using computa-
tional calculations. Theoretical studies on the reaction mecha-
nism of transition-metal-catalyzed cycloisomerizations of 1,6-
enynes have been reported.[18–25] Based on their studies, the
formation of bicyclo[4.1.0]heptenes from 1,6-enynes has al-
ready been established as a 6-endo-dig cyclization process.
Thus the primary aim of this work is not just pursuing the tra-
jectory of the reaction pathway toward bicyclo[4.1.0]heptenes
&
&
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
4
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
dynamic point of view and the generated [PdCl2(CH3CN)2ꢁn{P-
(OPh)3}n] (n=0, 1, 2) systems are probably in equilibrium with
each other. When these species are in contact with the 1,6-
enyne substrate, a catalyst-substrate adduct is either R1 ([Pd]=
[PdCl2{P(OPh)3}]) or R2 ([Pd]=[PdCl2{P(OPh)3]} or [PdCl2]) as
shown in Scheme 2. A detailed mechanism of the catalyst gen-
eration will be presented elsewhere.
When a catalyst–substrate adduct is a R2-type structure, it is
possible that [PdCl2{P(OPh)3}] or the [PdCl2] species can be an
active catalyst. However, the experimental results show that
the reaction is highly dependent on the nature of the addition-
al ligands, indicating that the active catalyst is not just a bare
[PdCl2] species (Table 2, entries 2–7). Moreover, the removal of
two CH3CN ligands from the [PdCl2(CH3CN)2] precursor requires
51.5 kcalmolꢁ1 of energy. Thus, we ruled out the case when
[PdCl2] is a catalyst species. On the other hand, the PdII species,
[PdCl2{P(OPh)3}2], may be transformed to the Pd0 species,
[Pd{P(OPh)3}2], in the presence of excess phosphite ligands.
Taking into account these features, the relative reaction ener-
getics of the several probable reaction pathways involving
these catalyst species, as shown in Scheme 2, were compared
to search for the most plausible reaction pathway as well as
the identity of the active catalyst species. Next, the effect of
the triphenylphosphine (PPh3) ligand on the catalyst per-
formance was investigated. To reproduce the experimental re-
sults, the catalyst systems were modeled without significant
deviations from the real structure. These catalysts were then
applied to (E)-N-(but-2-ynyl)-N-(3-cyclopropylallyl)-amine, which
resembles 2a except for the absence of the tosylate group at
the N tether, which was removed to avoid conformational
complexities. It is important to note that the substrate, (E)-N-
(but-2-ynyl)-N-(3-cyclopropylallyl)-amine, involves sophisticated
chemistry owing to the presence of a vinylcyclopropane
moiety. Thus limited but important pathways for the synthesis
of both Alder-ene and homo-ene type of products were con-
sidered in this study.
Figure 1. X-ray crystal structure of 2c.[17]
Computational details
The calculations were performed using a Gaussian 09 program
package.[26] All the results were obtained using density func-
tional theory (DFT) with B3LYP hybrid functional. “LANL2DZ+
ECP” basis set was used for the Pd atom, whereas the standard
6-31G(d) basis set was used for the remaining atoms. All the
geometries of stationary states were fully optimized without
symmetry. Solvation corrections were added using a self-con-
sistent reaction field (SCRF) approach by a conductor-like po-
larizable continuum model (CPCM). Toluene was used in the
CPCM calculations. Frequency calculations were performed to
characterize the stationary points as a minima or transition
states and to obtain the zero-point energies (ZPEs) and ther-
mal corrections at 298.15 K. The transition states were identi-
fied by using the single negative eigenvalue in the Hessian
matrix. Intrinsic reaction coordinate calculations (IRC) were per-
formed to ensure that the transition states connect the pro-
posed reactants, intermediates, or products.
Scheme 2. A plausible reaction mechanism for Pd-catalyzed cycloisomeriza-
tions.
but deciphering why the catalytic system, particularly [PdCl2-
(MeCN)2]/P(OPh)3, prefers such a pathway over one toward
homo-ene product. The detailed reaction paths investigated in
this work are summarized in Scheme 2.
The calculated energies of [PdCl2(CH3CN){P(OPh)3}] and
[PdCl2{P(OPh)3}2] relative to the [PdCl2(CH3CN)2] precursor are
ꢁ1.8 and ꢁ1.5 kcalmolꢁ1 in toluene, respectively, whereas that
of [PdCl2{P(OPh)3}] is 1.5 kcalmolꢁ1. The exchange of CH3CN li-
gands with P(OPh)3 appears to be smooth in terms of thermo-
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Consideration of the reaction pathway using the
[PdCl2{P(OPh)3}] catalyst
nal-bipyramidal structure, in which two Clꢁ ligands are trans to
each other in the axial positions, and three other ligands are in
equatorial positions. Given the strong preference of Pd metal
to possess a 16-electron square-planar geometry, an 18-elec-
tron trigonal-bipyramidal [PdCl2{P(OPh)3}(h2-ethyne)(h2-ethyl-
ene)] geometry may be less favorable than a square-planar
[PdCl2{P(OPh)3}(h2-ethyne)] geometry. Therefore, the formation
of metallacyclopentene intermediate (P4_POCl2) through con-
ventional oxidative cyclization from the h4-coordinated reac-
tant state is not possible using the [PdCl2{P(OPh)3}] catalyst.
However, the optimization of P4_POCl2 was successful. P4_
POCl2 does not exhibit a trigonal-bipyramidal structure but
rather exhibits a square-pyramidal structure in which the alken-
yl carbon is at the apex position. Interestingly, the pathway to
generate P4_POCl2 from the h2-coordinated R1_POCl2 was
fortuitously found. This reaction is slightly endothermic
(2.7 kcalmolꢁ1) with an activation energy barrier of 40.8 kcal
molꢁ1. The energy barrier is quite large, and thus cannot be
easily overcome by the conventional reaction conditions. Alter-
natively, P4_POCl2 can be generated from syn-P5B_POCl2.
The anti-to-syn isomerization of P5_POCl2 to P5B_POCl2
occurs with a rotational energy barrier of 12.8 kcalmolꢁ1. In ad-
dition, the subsequent rearrangement by the cleavage of the
inner cyclopropyl ring requires an activation energy barrier of
The complexation of the [PdCl2{P(OPh)3}] catalyst to the ethyne
group of the substrate (R1_POCl2) in h2 fashion leads to the
intramolecular nucleophilic attack of the alkene group through
either 5-exo-dig or 6-endo-dig cyclization (Figure 2). Whereas
the 5-exo-dig process is mildly endothermic by 0.6 kcalmolꢁ1
the 6-endo-dig cyclization process is exothermic by 1.2 kcal
molꢁ1 (P6_POCl2).
29.3 kcalmolꢁ1
.
b-Hydrogen elimination of metallacycles has been known to
be regiospecific.[2a] In general, b-hydrogen elimination of Ha is
geometrically more favorable, but less favorable energetically,
leading to the competition between the formations of 1,3-
diene and 1,4-diene (Scheme 3). However, a branching at the
allylic position shifts the regioselectivity toward the 1,3-dien-
e.[2a,4a,c] The experimental results in this study show that the vi-
nylcyclopropyl moiety preferably leads to the formation of
Figure 2. Free-energy reaction profile of Pd-catalyzed 5-exo-dig, 6-endo-dig,
and oxidative cyclizations of (E)-N-(but-2-ynyl)-N-(3-cyclopropylallyl)amine in
toluene as solvent. [Pd]=[PdCl2{P(OPh)3}]. Numbers in red are DG at 2988C
(energies in kcalmolꢁ1).
Alder-ene product (1,4-diene) when
a
typical catalyst,
The activation energy barriers of these processes are
18.6 kcalmolꢁ1 (TS_R1_P5_POCl2) and 17.6 kcalmolꢁ1 (TS_R1_
P6_POCl2) for the former and latter, respectively. Although
TS_R1_P6_POCl2 is slightly favorable, the energy difference
between these two pathways is insignificant. The dominant
production of 3-aza-bicyclo[4.1.0]heptene observed in the ex-
periment indicates that the reaction pathway cannot be deter-
mined from the first step.
[Pd(PPh3)4], was used. Thus, calculations based on “b-hydrogen
elimination of Ha followed by reductive elimination” were per-
formed. When [PdCl2{P(OPh)3}] was used as the catalyst, the
It has been known that the Alder-ene type of products can
be formed by b-hydrogen elimination from the metallacyclo-
pentene intermediate. The oxidative cyclization of the R2-type
reactant state, as shown in Scheme 2, has been proposed as
the major route toward the metallacyclopentene intermedia-
te.[2a] However, the complexation of [PdCl2{P(OPh)3}]
to both the ethyne and ethylene groups of the sub-
strate in h4 fashion was unsuccessful during the cal-
culations; [PdCl2{P(OPh)3}] cannot simultaneously co-
ordinate to both the ethyne and the ethylene
groups. Geometry optimization yielded the formation
of either an ethyne-bound or an ethylene-bound
structure. R1_POCl2 exhibit a square-planar structure,
Scheme 3. Two major pathways of b-H elimination.[2a]
Scheme 4. Formation of Alder-ene product via a metallacyclopentadiene intermediate.
[Pd]=[PdCl2{P(OPh)3}]. Numbers in parentheses are DG values at 2988C (energies in kcal
in which a triphenylphosphite ligand, P(OPh)3, is trans
to the ethyne group. R2_POCl2 may possess a trigo- molꢁ1).
&
&
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
6
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 3. Free-energy reaction profile towards the Alder-ene and homo-ene product from the metallacyclopentene intermediate in toluene solvent. The
direct conversion from P5B to the homo-ene product is shown by the purple line. [Pd]=[PdCl2{P(OPh)3}]. Numbers in red are DG at 2988C (energies in kcal
molꢁ1).
b-hydrogen elimination product (P3_POCl2) was found to be
highly unstable (27.5 kcalmolꢁ1, Scheme 4 and Figure 3). The
reaction is endothermic by 24.8 kcalmolꢁ1 and the activation
energy barrier is 26.7 kcalmolꢁ1. The generated metal-hydride
intermediate undergoes reductive elimination with an energy
barrier of 8.9 kcalmolꢁ1 (TS_P3_P3D_POCl2).
The b-hydrogen elimination followed by reductive elimina-
tion, as shown in Scheme 4, can afford an Alder-ene product;
however, it does not explain the formation of homo-ene prod-
uct (2c) in which the cyclopropane cleavage plays an impor-
tant role. Therefore, we searched for a pathway that leads to
the formation of such a product. In the cyclization of 1,6-
enynes containing a vinylcyclopropane moiety, the rearrange-
ment leading to the metallacyclooctadiene species has been
reported for Rh,[27] Ni,[28] Fe,[29] and Ru,[30] and the detailed
mechanistic pathways by using computational calculations
have been investigated (Scheme 5).[31]
Scheme 5. Formation of homo-ene product via metallacyclopentadiene !
metallacyclooctadiene intermediates. [Pd]=[PdCl2{P(OPh)3}]. Numbers in pa-
When the [PdCl2{P(OPh)3}] catalyst is used, it is prerequisite
rentheses are DG values at 2988C (energies in kcalmolꢁ1).
for the P(OPh)3 ligand to move to the trans position of the
vinyl carbon. The resulting P4B_POCl2 system has a similar
square-pyramidal structure, but the ethynyl carbon is now situ-
ated at the apex position. This isomerization is thermodynami-
cally uphill by 5.5 kcalmolꢁ1. The following rearrangement
toward the metallacyclooctadiene is shown to be further endo-
thermic by 3.4 kcalmolꢁ1 (P7_POCl2) and the activation energy
barrier for this process is substantial (31.2 kcalmolꢁ1). P7_
POCl2 undergoes b-hydrogen elimination to form the P2_
POCl2 intermediate, which leads to P2D_POCl2 through a hy-
drogen transfer. The activation energy barrier for the b-hydro-
gen elimination is again very large (28.9 kcalmolꢁ1). At the P7_
POCl2 state, both the forward reaction to P2_POCl2 and the
backward reaction to P4B_POCl2 require similarly large activa-
tion barriers near 30 kcalmolꢁ1 and thus the reaction might be
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
7
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
retarded at this stage. Once the reaction proceeds to P2_
POCl2, the activation barrier for the forward reaction to P2D_
POCl2 is only 7.6 kcalmolꢁ1 due to the instability of P2_POCl2
state (21.7 kcalmolꢁ1), whereas that for the P2_POCl2!P7_
POCl2 backward reaction is 18.8 kcalmolꢁ1. Thus, the forma-
tion of thermodynamically stable P2D_POCl2 from P2_POCl2
appears to be smooth.
(Figure 4). The activation energy barriers for these processes
are 32.5 and 30.3 kcalmolꢁ1 for the former and the latter, re-
spectively, which are more than 10 kcalmolꢁ1 larger than those
using the [PdCl2{P(OPh)3}] catalyst. In addition, the metallacy-
clopentene pathway from the h4-coordinated reactant state
was not a viable process; we were not able to locate the ex-
pected R2_PO2 structure similar to the case using [PdCl2{P-
(OPh)3}] catalyst. Instead, the optimization process from the
structure, in which [Pd{P(OPh)3}2] was coordinated to both the
ethyne and ethylene groups of the substrate, resulted in the
formation of R2_PO with one phosphite ligand being removed
from the Pd center. The size of the groove encompassed by
the substrate is not spacious enough to accommodate two
bulky P(OPh)3 ligands, which may have caused the loss of one
phosphite ligand from the Pd center.
Interestingly, we could locate the transition state that leads
P5B_POCl2 directly to P2_POCl2 through the concomitant
CꢁC bond cleavages of the two cyclopropanes and the b-hy-
drogen elimination (Scheme 6). The overall energy barrier of
this process is lower (TS_P5B_P2_POCl2, 35.3 kcalmolꢁ1) than
that of the pathway through the metallacyclooctadiene (TS_
P7_P2_POCl2, 40.5 kcalmolꢁ1). However, even the energy of
TS_P5B_P2_POCl2 is twice as large as the overall energy barri-
er of the 6-endo-dig cyclization.
Similar to the case with [Pd{P(OPh)3}2], both the 5-exo-dig
and 6-endo-dig cyclizations from R1_PO, in which
a
[Pd{P(OPh)3}] catalytic moiety is coordinated to the ethyne
group of the substrate in h2 fashion, were found to be highly
endothermic. The 6-endo-dig cyclization requires large amount
of activation energy barrier (28.7 kcalmolꢁ1), and the resulting
P6_PO is 16.9 kcalmolꢁ1 higher in energy compared with its re-
actant state. This unstable intermediate prompts to the stable
P6D_PO by the hydrogen shift. The activation energy barrier
for the 5-exo-dig process is even larger (30.2 kcalmolꢁ1), and
the energy of P5_PO is 21.9 kcalmolꢁ1. P5_PO is likely in equi-
librium with its rotational isomer P5B_PO, from which the for-
mation of stable metallacyclopentene intermediate (P4_PO) is
easily accessible. P5B_PO is similar in energy as P5_PO
(21.1 kcalmolꢁ1), and the rotational energy barrier between
Scheme 6. Formation of homo-ene product directly from P5B_POCl2.
[Pd]=[PdCl2{P(OPh)3}]. Numbers in parentheses are DG at 2988C (energies
in kcalmolꢁ1).
Houk and co-workers reported that the theoretically calcu-
lated activation energy barriers to form a metallacyclooctadiene
intermediate from a metallacyclopentene derivative are very
mild (1–2 kcalmolꢁ1), when the O-tethered 1,6-enynes with Niꢁ P5_PO and P5B_PO is about 6–7 kcalmolꢁ1. The Gibbs free
NHC catalyst or C-tethered 1,6-enynes with RuꢁCp catalyst
were considered.[31] However, a similar process for N-tethered
1,6-enynes using the [PdCl2{P-
energy of TS_P4_P5B_PO is 37.3 kcalmolꢁ1, which is about
7 kcalmolꢁ1 higher than that of TS_R1_P5_PO. The formation
(OPh)3}] catalyst requires a sub-
stantial amount of activation
energy barrier, indicating that
the pathway toward the homo-
ene product through metallacy-
clooctadiene intermediate is not
favored in the presence of the
[PdCl2{P(OPh)3}] catalyst.
Consideration of the reaction
pathway using [Pd{P(OPh)3}n]
catalysts
Next, the possibility of the
[Pd{P(OPh)3}2] species as the
active catalyst was evaluated.
Surprisingly, both 5-exo-dig and
6-endo-dig cyclization products
are unstable when [Pd{P(OPh)3}2]
is used; the energies of P5_PO2
and P6_PO2 products relative
Figure 4. Free-energy reaction profile of Pd-catalyzed 5-exo-dig, 6-endo-dig, and oxidative cyclizations of (E)-N-
(but-2-ynyl)-N-(3-cyclopropylallyl)amine in toluene solvent. [Pd]=[Pd{P(OPh)3}2] for PO2, [Pd{P(OPh)3}] for PO.
Numbers in red are DG at 2988C (energies in kcalmolꢁ1).
to the R1_PO2 reactant are 18.8
and 16.4 kcalmolꢁ1, respectively
&
&
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
8
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 5. Free-energy reaction profile toward Alder-ene and homo-ene products from the metallacyclopentene intermediate in toluene solvent.
[Pd]=[Pd{P(OPh)3}]. Numbers in red are DG at 2988C (energies in kcalmolꢁ1).
of P4_PO from P5B_PO requires 16.2 kcalmolꢁ1 of activation
energy; thus, it is expected that the reaction toward P4_PO is
not favored compared with that toward P6D_PO.
occurs very smoothly. This process is endothermic by 4.1 kcal
molꢁ1 similar to the case using [PdCl2{P(OPh)3}]; however, the
activation energy barrier is only 11.2 kcalmolꢁ1. It should be
noted that the same process with [PdCl2{P(OPh)3}] catalyst has
an activation energy barrier of 31.2 kcalmolꢁ1. The following b-
hydrogen elimination step is also facilitated compared to that
for [PdCl2{P(OPh)3}] counterpart; it is exothermic by 4.9 kcal
molꢁ1 with an activation energy barrier of 15.5 kcalmolꢁ1. It
should be noted that the same process using [PdCl2{P(OPh)3}]
is 10.1 kcalmolꢁ1 endothermic, and the activation energy barri-
er is 28.9 kcalmolꢁ1. The direct route to P2_PO from the P5B_
PO product state is also possible. However, the energy level of
TS_P5B_P2_PO is 51.0 kcalmolꢁ1, which is too high to over-
come; therefore, this pathway could be ignored.
On the other hand, the direct formation of P4_PO from R2_
PO by oxidative metallacyclization is possible through TS_R2_
P4_PO. The energy level of TS_R2_P4_PO is 39.7 kcalmolꢁ1,
which is only 2.4 kcalmolꢁ1 higher than that of TS_P4_P5B_
PO. Thus, the formation of metallacyclopentene intermediate
(P4_PO) using [Pd{P(OPh)3}] catalyst is perceived as occurring
through either the 5-exo-dig cyclization from h2-coordinated
R1_PO or the direct oxidative cyclization from h4-coordinated
R2_PO, with the former being slightly more favorable. This
result is unprecedented because the formation of palladacyclo-
pentene intermediate has been assumed to occur only
through the oxidative cyclization from the R2-type reactant.
P4_PO undergoes b-hydrogen elimination with an activation
energy barrier of 29.1 kcalmolꢁ1 (Figure 5). The resultant P3_
PO is unstable (21.5 kcalmolꢁ1), similar to P3_POCl2. Contrary
to the case of P3_POCl2, in which the activation energy barrier
for further reductive elimination is larger (8.9 kcalmolꢁ1) than
that of the backward reaction to P4_POCl2 (1.9 kcalmolꢁ1), the
reductive elimination process for P3_PO is barrierless, indicat-
ing that the formation of Alder-ene product is facile when the
Pd0L1 catalyst is used.
The results described above show that the formation of
bicyclo[4.1.0]heptenes is still favored over either the Alder-ene
or homo-ene product when [Pd{P(OPh)3}] was used as the cata-
lyst. However, it should be noted that the pathway toward
homo-ene product is much more facilitated once the metalla-
cyclopentene intermediate is formed. In addition, it would be
reasonable that bicyclo[4.1.0]heptenes and homo-ene product
are somewhat in competition when [Pd{P(OPh)3}] was used as
the catalyst.
The entire reaction profiles of [PdCl2{P(OPh)3}] and [Pd{P-
(OPh)3}] catalysts are compared in Figure 6. The most probable
reaction pathway using [PdCl2{P(OPh)3}] catalyst consists of the
6-endo-dig cyclization followed by a hydrogen shift leading to
P6D with an overall activation energy of 17.6 kcalmolꢁ1. When
The P4B_PO state (5.7 kcalmolꢁ1), in which the P(OPh)3
ligand has migrated, is almost isoenergetic with P4_PO
(4.8 kcalmolꢁ1), and the subsequent cyclopropane cleavage
process toward metallacyclooctadiene intermediate (P7_PO)
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
9
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
Figure 6. Free-energy reaction profile of [PdCl2{P(OPh)3}]- (black) and [Pd{P(OPh)3}]-(gray) catalyzed cycloisomerization of (E)-N-(but-2-ynyl)-N-(3-cyclopropyl-
allyl)amine in toluene solvent. Numbers are DG at 2988C (energies in kcalmolꢁ1).
the catalyst species is [Pd{P(OPh)3}], a pathway leading to
P6D_PO is still more favorable over P2D_PO but the process
from P4_PO to P2_PO is very smooth compared with the case
of [PdCl2{P(OPh)3}]; a mixture of both P2D_PO and P6D_PO as
final products is expected using this catalyst. Therefore, the
possibility that [PdCl2{P(OPh)3}] is an active catalyst is more
likely according to the calculation results.
detail, the reaction profiles catalyzed by [PdCl2(PPh3)] and
[PdCl2{P(OPh3)}] are compared, as shown in Figure 7, whereas
those by [Pd(PPh3)] and [Pd{P(OPh3)}] are compared, as shown
in Figure 8. As shown in the two Figures, the energy profiles of
the reactions catalyzed by [PdCl2(PPh3)] (black lines in Figure 7)
or [Pd(PPh3)] (gray lines in Figure 8) are generally higher than
those of [PdCl2{P(OPh3)}] (gray lines in Figure 7) or [Pd{P(OPh3)}]
(black lines in Figure 8) implying that the production of both
3-aza-bicyclo[4.1.0]heptene and homo-ene products are disfa-
vored and thus the yields of both products should decrease.
However, the experimental results showed a significant de-
crease in the yield of 2c and an increased yield of 2b (Table 2,
entry 3), which needs an alternate explanation.
Cases with the PPh3 ligand: [PdCl2(PPh3)] and [Pd(PPh3)]
catalysts
Next, the reason why P(OPh)3 ligand, in particular, facilitates
the 6-endo-dig cyclization was evaluated. The experimental re-
sults showed that the reaction catalyzed by the [PdCl2-
(CH3CN)2]/PAr3 catalyst systems gave not only the 3-aza-
bicyclo[4.1.0]heptene (2c), but also the homo-ene products
(2b) as well, even though the relative ratios between these
two products vary by ligands (Table 2). Moreover, the yields of
2c were significantly decreased, whereas those of 2b were in-
creased using [PdCl2(CH3CN)2]/PAr3 catalyst systems. Thus, the
calculations were performed with [PdCl2(PPh3)] and [Pd(PPh3)],
which resemble [PdCl2{P(OPh3)}] and [Pd{P(OPh3)}], respectively,
and thus a linear comparison between the two ligand systems
was possible. Moreover, this study may provide a trend for the
reactions catalyzed by PdII and Pd0 species, which is important
to determine the active catalyst species.
The Gibbs free energies of activation energy barrier for each
step of the reaction pathways are listed in Table 4. When
[PdCl2{P(OPh3)}] is used as the catalyst, the maximum activation
energy barrier of the most efficient reaction pathway leading
to the homo-ene product is 33.9 kcalmolꢁ1 for the P5B!P7
process; the overall reaction pathway is thus R1!P5!P5B!
P2!P2D. This value is twice as large as that for the overall
barrier of R1!P6D process. Therefore the production of P6D
is dominantly favored. When [PdCl2(PPh3)] is used as the cata-
lyst, the reaction pathway leading to P2D is as follows: R1!
P5!P5B!P4!P4B!P7!P2!P2D. The maximum activa-
tion energy barrier in this case is 27.6 kcalmolꢁ1 for P5B!P4
process. This value is only 7.4 kcalmolꢁ1 larger than that for
the overall barrier of R1!P6D process. Thus, the formation of
P2 and P6D should be in competition, with the latter being
more favorable.
The overall reaction patterns catalyzed by [PdCl2(PPh3)] and
[Pd(PPh3)] are similar to those catalyzed by [PdCl2{P(OPh3)}]
and [Pd{P(OPh3)}], respectively. To analyze the results in more
&
&
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
10
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
Figure 7. Free-energy reaction profile of [PdCl2{P(OPh)3}] (black) and [PdCl2(PPh)3] (gray) catalyzed cycloisomerization of (E)-N-(but-2-ynyl)-N-(3-cyclopropyl-
allyl)amine in toluene solvent. Numbers are DG at 2988C (energies in kcalmolꢁ1).
Figure 8. Free-energy reaction profile of [Pd{P(OPh)3}]- (gray) and [Pd(PPh)3]- (black) catalyzed cycloisomerization of (E)-N-(but-2-ynyl)-N-(3-cyclopropylallyl)-
amine in toluene solvent. Numbers are DG at 2988C (energies in kcalmolꢁ1).
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
11
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
to be 17.6 kcalmolꢁ1. This amount is relatively larger
than those reported for Pt (9–13 kcalmolꢁ1)[7c,18,20]
and Au (4–6 kcalmolꢁ1)[8c,18] indicating that harsher
reaction conditions are necessary for the Pd catalyst.
The calculation results provide interesting point that
the pathway involving metallacycle intermediate re-
quires a very high activation barrier for the Pd cata-
lyst species (40–42 kcalmolꢁ1) similar to Pt catalysts
(30–35 kcalmolꢁ1).[7c,20d] This result is in contrast to
Table 4. Gibbs free energies of activation barrier for each step of the reaction path-
ways catalyzed by four different hypothetical catalyst species.
DGꢀ [kcalmolꢁ1
Homo-ene
]
Catalyst
6-endo-dig
R1! P5! P5B! P4! P7! P5B! P2! R1! P6!
P5
P5B
P4
P7
P2
P2
P2D
P6
P6D
[PdCl2{P(OPh3)}] 18.6 12.8
29.3
36.7 28.9 33.9[a] 7.6
17.6 5.9
20.2 6.9
28.7 6.2
31.5 7.2
[PdCl2(PPh3)]
[Pd{P(OPh3)}]
[Pd(PPh3)]
21.5 11.8
30.2[a] 6.1
27.6[a] 23.1 25.1 35.4
5.3
2.3
1.8
16.2
18.7
12.1 15.5 28.1
14.9 16.7 29.2
32.6[a]
5
the cases of Ni[31b] (ꢀ20 kcalmolꢁ1
) (
or Ru[31c]
ꢀ13 kcalmolꢁ1) catalysts. Selective formation of
Alder-ene type products reported previously may in-
volve other pathways or other catalyst species as
[a] Maximum activation-energy barrier during the pathway toward P2D.
This interpretation is in excellent agreement with the experi-
mental result. Such an analysis clearly shows why Pd0 species,
for example, [Pd{P(OPh3)}] or [Pd(PPh3)], cannot be the active
catalyst. The reaction pathway leading to P2D is R1!P5!
P5B!P4!P4B!P7!P2!P2D, irrespective of the type of
Pd0 catalyst species used. When [Pd{P(OPh3)}] is used as the
catalyst, the maximum activation energy barrier toward homo-
ene product (P2D) is 30.2 kcalmolꢁ1 for R1!P5 process, which
is comparable to the overall activation energy barrier for R1!
P6D process (28.7 kcalmolꢁ1). The dominant production of 3-
aza-bicyclo[4.1.0]heptane cannot be explained with such a sce-
nario. The same is true for the [Pd(PPh3)] case. The maximum
activation energy barrier toward P2D is 32.6 kcalmolꢁ1 for the
R1!P5 process, whereas that toward P6D is 31.5 kcalmolꢁ1,
which again indicates that 3-aza-bicyclo[4.1.0]heptane cannot
be a dominant product. Therefore, it is reasonable to conclude
that the unprecedented production of bicyclo[4.1.0]heptenes
from 1,6-enynes is the results of cycloisomerization catalyzed
by PdII species, [PdCl2{P(OPh3)}].
well, which deserves further investigation.
The current efforts are directed at gaining a deeper under-
standing of some of the fundamental principles dictating the
reactivity of the substrates, as well as extending their utility by
expanding the substrate scope. Further experiments have to
be carried out to corroborate the proposed mechanism. We
envision that the reaction developed in this study will be easily
accessible to many synthetic chemists and of use to them in
near future.
Experimental Section
Experimental details
To a 20 mL tube-type flask [PdCl2(MeCN)2] (0.03 mmol), triphenyl
phosphite (0.12 mmol), and toluene (2 mL) were added and
capped with a rubber septum. After the solution was stirred for
5 min at room temperature, a substrate (0.3 mmol) dissolved in tol-
uene (1 mL) was added to the reactor through a syringe. The re-
sulting solution was heated in a thermostat-controlled oil bath
(1308C). The reaction mixture was stirred until the substrate was
completely consumed (reaction monitored by TLC). After the sol-
vent was removed by a rotary evaporator, the residue was purified
by flash column chromatography on a silica gel column by eluting
with hexane/ethyl acetate (15:1).
Conclusion
We have developed a Pd-catalyzed cycloisomerization reaction
of N- and O-bridged 1,6-enynes to afford 3-aza- and 3-
oxabicyclo[4.1.0]heptenes in reasonable to high yields. The re-
actions were catalyzed by [Pd(maleimidate)2(PPh3)2] or [PdCl2-
(CH3CN)2]/P(OPh)3. The computational calculations using DFT
strongly indicate that [PdCl2{P(OPh3)}] in the PdII oxidation
state rather than [Pd{P(OPh3)}n] in the Pd0 oxidation state is the
active catalyst species responsible for the formation of 3-aza-
bicyclo[4.1.0]heptenes. The designation of [PdCl2{P(OPh3)}] as
an active catalyst species has not been considered yet and
thus highlights the importance of the calculation results ob-
tained in this work. This catalyst species transforms a model
substrate, (E)-N-(but-2-ynyl)-N-(3-cyclopropylallyl)-amine, to 7-
Acknowledgements
This research was supported by Basic Science Research Pro-
gram through the National Research Foundation (NRF) funded
by the Ministry of Education, Science and Technology (2007–
0093864, 2012–0001691, 2012R1A2A2A01013004). This re-
search was also supported by the PLSI supercomputing resour-
ces of Korea Institute of Science and Technology Information.
HMO and JHK thanks the Brain Korea 21 for fellowships.
cyclopropyl-6-methyl-3-azabicyclo[4.1.0]hept-4-ene
through
Keywords: enynes · cyclization · density functional theory ·
isomerization · palladium
the 6-endo-dig cyclization pathway from the h2-coordinated re-
actant state (R1). The overall activation barrier of this process
is far lower than that toward Alder-ene or homo-ene products.
For the formation of homo-ene product, a direct double cyclo-
propyl cleavage reaction from P5B to P2 is slightly more favor-
able than a pathway through metallacyclopentene (P4)!met-
allacyclooctene (P7) intermediates. The reaction barrier for the
6-endo-dig cyclization pathway studied in this work was found
[1] a) N. Krause, O. Aksin-Artok, M. Asikaine, M. V. Breker, C. Deutsch, J. Erd-
sack, H. T. Fan, B. Gockel, S. Minkler, M. Poonoth, Y. Sawama, T. Sun, F.
Volz, C. Winter, J. Organomet. Chem. 2012, 704, 1–8; b) P. Y. Toullec, V.
Michelet in Topics in Current Chemistry, Vol. 302, Springer, Berlin, 2011,
pp. 31–80; c) C. Bruneau, Angew. Chem. 2005, 117, 2380–2386; Angew.
Chem. Int. Ed. 2005, 44, 2328–2334; d) C. Aubert, O. Buisine, M. Malac-
ria, Chem. Rev. 2002, 102, 813–834.
&
&
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
12
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Paper
[2] a) C. Aubert, O. Buisine, M. Malacria, Chem. Rev. 2002, 102, 813–834;
b) M. Mori in Topics in Organometallic Chemistry, Vol. 1, Springer, Berlin,
1998, pp. 133–154; c) C. S. Poulsen, R. Madsen, Synthesis 2003, 1–8;
d) B. M. Trost, M. J. Krische, Synlett 1998, 1–16; e) B. M. Trost, M. Yanni,
K. Hoogsteen, J. Am. Chem. Soc. 1993, 115, 5294–5295; f) B. M. Trost,
F. D. Toste, J. Am. Chem. Soc. 2002, 124, 5025–5036; g) L. Ye, Q. Chen, J.
Zhang, V. Michelet, J. Org. Chem. 2009, 74, 9550–9553; h) Y. T. Lee, Y. K.
Kang, Y. K. Chung, J. Org. Chem. 2009, 74, 7922–7934.
[3] a) J. Tsuji in Palladium Reagents and Catalysts: New Perspectives for the
21st Century, Wiley, Chichester, 2004; b) E. Negishi, A. de Meijre in Orga-
nopalladium Chemistry for Organic Synthesis, Wiley, New York, 2002; c) J.
Hartwig in Organotransition Metal Chemistry, University Science Books,
Sausalito, 2010; d) N. Selander, K. J. Szabꢁ, Chem. Rev. 2011, 111, 2048–
2076.
[4] a) B. M. Trost, M. Lautens, J. Am. Chem. Soc. 1985, 107, 1781–1783;
b) B. M. Trost, G. J. Tanoury, J. Am. Chem. Soc. 1988, 110, 1636–1638;
c) B. M. Trost, Acc. Chem. Res. 1990, 23, 34–42; d) B. M. Trost, M. Laut-
ens, M. C. Chan, D. J. Jebaratnam, T. Mueller, J. Am. Chem. Soc. 1991,
113, 636–644; e) B. M. Trost, O. J. Gelling, Tetrahedron Lett. 1993, 34,
8233–8236; f) B. M. Trost, G. J. Tanoury, M. Lautens, C. Chan, D. T. Mac-
Pherson, J. Am. Chem. Soc. 1994, 116, 4255–4267; g) B. M. Trost, R. I. Hi-
guchi, J. Am. Chem. Soc. 1996, 118, 10094–10105.
[5] a) V. Michelet, P. Y. Toullec, J. P. GenÞt, Angew. Chem. 2008, 120, 4338–
4386; Angew. Chem. Int. Ed. 2008, 47, 4268–4315; b) J. C. Galland, M.
Savignac, J. P. GenÞt, Tetrahedron Lett. 1997, 38, 8695–8698; c) J. C. Gal-
land, S. Dias, M. Savignac, J. P. GenÞt, Tetrahedron 2001, 57, 5137–5148;
d) L. Charruault, V. Michelet, J.-P. GenÞt, Tetrahedron Lett. 2002, 43,
4757–4760.
[14] I. J. S. Fairlamb, A. R. Kapdi, J. M. Lynam, R. J. K. Taylor, A. C. Whitwood,
Tetrahedron 2004, 60, 5711–5718.
[15] a) N. Mungwe, A. J. Swarts, S. F. Mapolie, G. Westman, J. Organomet.
Chem. 2011, 696, 3527–3535; b) M. L. Kantam, P. V. Reddy, P. Srinivas, S.
Bhargava, Tetrahedron Lett. 2011, 52, 4490–4493; c) E. E. Lee, R. A.
Batey, Angew. Chem. 2004, 116, 1901–1904; Angew. Chem. Int. Ed. 2004,
43, 1865–1868; d) R. Ghosh, N. N. Adarsh, A. Sarkar, J. Org. Chem. 2010,
75, 5320–5322; e) M. Lautens, J. L. Renaud, J. Hiebert, J. Am. Chem. Soc.
2000, 122, 1804–1805.
[16] A. Fꢂrstner, H. Szillat, F. Stelzer, J. Am. Chem. Soc. 2000, 122, 6785–
6786.
[17] CCDC-985746 (2c) and CCDC-985747 (15c) contain the supplementary
crystallographic data for this paper. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif.
[18] M. P. MuÇoz, M. Mꢃndez, C. Nevada, D. J. Cꢄrdenas, A. M. Echavarren,
Chem. Eur. J. 2003, 9, 2627–2635.
[19] C. Nevado, L. Charruault, V. Michelet, C. Nieto-Oberhuber, M. P. MuÇoz,
M. Mꢃndez, M. N. Rager, J. P. GenÞt, A. M. Echavarren, Eur. J. Org. Chem.
2003, 706–713.
[20] a) E. Soriano, P. Ballesteros, J. Marco-Contelles, J. Org. Chem. 2004, 69,
8018–8023; b) E. Soriano, J. Marco-Contelles, J. Org. Chem. 2005, 70,
9345–9353; c) E. Soriano, P. Ballesteros, J. Marco-Contelles, Organome-
tallics 2005, 24, 3172–3181; d) E. Soriano, P. Ballesteros, J. Marco-Con-
telles, Organometallics 2005, 24, 3182–3191.
[21] a) B. M. Trost, A. S. K. Hashmi, Angew. Chem. 1993, 105, 1130–1132;
Angew. Chem. Int. Ed. Engl. 1993, 32, 1085–1087; b) B. M. Trost, A. S. K.
Hashmi, J. Am. Chem. Soc. 1994, 116, 2183–2184; c) B. M. Trost, G. A.
Doherty, J. Am. Chem. Soc. 2000, 122, 3801–3810; d) B. M. Trost, D. F.
Toste, A. B. Pinkerton, Chem. Rev. 2001, 101, 2067–2096.
[6] a) Y. Yamamoto, S. Kuwabara, Y. Ando, H. Nagata, H. Nishiyama, K. Itoh,
J. Org. Chem. 2004, 69, 6697–6705; b) S. C. Zhao, K. G. Ji, L. Lu, T. He,
A. X. Zhou, R. L. Yan, S. Ali, X. Y. Liu, Y. M. Liang, J. Org. Chem. 2012, 77,
2763–2772.
[22] N. Chatani, K. Kataoka, S. Murai, N. Furokawa, Y. Seki, J. Am. Chem. Soc.
1998, 120, 9104–9105.
[7] a) J. Blum, H. Beer-Kraft, Y. Badrieh, J. Org. Chem. 1995, 60, 5567–5569;
b) A. Fꢂrstner, F. Stelzer, H. Szillat, J. Am. Chem. Soc. 2001, 123, 11863–
11869; c) M. Mꢃndez, M. P. MuÇoz, C. Nevado, D. J. Cꢄrdenas, A. M. Echa-
varren, J. Am. Chem. Soc. 2001, 123, 10511–10520.
[23] a) B. M. Trost, M. K. Trost, J. Am. Chem. Soc. 1991, 113, 1850–1852;
b) B. M. Trost, V. K. Chang, Synthesis 1993, 824–832.
[24] B. M. Trost, A. S. K. Hashmi, R. G. Ball, Adv. Synth. Catal. 2001, 343, 490–
494.
[8] a) For the first gold-catalyzed cycloisomerization of a substrate with an
enyne-substructure, see: A. S. K. Hashmi, T. M. Frost, J. W. Bats, J. Am.
Chem. Soc. 2000, 122, 11553–11554; b) for a general review on gold
catalysis, see: A. S. K. Hashmi, G. J. Hutchings, Angew. Chem. 2006, 118,
8064–8105; Angew. Chem. Int. Ed. 2006, 45, 7896–7936; c) Nieto-Ober-
huber, M. P. MuÇoz, E. BuÇuel, C. Nevado, D. J. Cꢄrdenas, A. M. Echavar-
ren, Angew. Chem. 2004, 116, 2456–2460; Angew. Chem. Int. Ed. 2004,
43, 2402–2406.
[25] K. Mikami, M. Hatano, Proc. Natl. Acad. Sci. USA 2004, 101, 5767–5769.
[26] Gaussian 09, Revision A.02, M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E.
Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, B. Men-
nucci, G. A. Petersson, H. Nakatsuji, M. Caricato, X. Li, H. P. Hratchian,
A. F. Izmaylov, J. Bloino, G. Zheng, J. L. Sonnenberg, M. Hada, M. Ehara,
K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O.
Kitao, H. Nakai, T. Vreven, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro,
M. Bearpark, J. J. Heyd, E. Brothers, K. N. Kudin, V. N. Staroverov, R. Ko-
bayashi, J. Normand, K. Raghavachari, A. Rendell, J. C. Burant, S. S. Iyen-
gar, J. Tomasi, M. Cossi, N. Rega, J. M. Millam, M. Klene, J. E. Knox, J. B.
Cross, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E. Stratmann,
O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W. Ochterski, R. L. Martin,
K. Morokuma, V. G. Zakrzewski, G. A. Voth, P. Salvador, J. J. Dannenberg,
S. Dapprich, A. D. Daniels, ꢅ. Farkas, J. B. Foresman, J. V. Ortiz, J. Cio-
slowski, D. J. Fox, Gaussian, Inc., Wallingford CT, 2009.
[27] P. A. Wender, H. Takahashi, B. Witulski, J. Am. Chem. Soc. 1991, 113,
1850–1852.
[28] G. Zuo, J. Louie, J. Am. Chem. Soc. 2005, 127, 5798.
[29] A. Fꢂrstner, K. Majima, R. Martin, H. Krause, E. Kattnig, R. Goddard, C. W.
Lehmann, J. Am. Chem. Soc. 1991, 113, 1850–1852.
[30] a) B. M. Trost, F. D. Toste, H. C. Shen, J. Am. Chem. Soc. 2000, 122, 2379;
b) B. M. Trost, H. C. Shen, D. B. Horne, E. D. Toste, B. G. Steinmetz, C. Kor-
adin, Chem. Eur. J. 2005, 11, 2577.
[9] E. Benedetti, A. Simonneau, A. Hours, H. Amouri, A. Penoni, G. Palmisa-
no, M. Malacria, J. P. Goddard, L, Fensterbank, Adv. Synth. Catal. 2011,
353, 1908–1912.
[10] a) S. Y. Kim, Y. K. Chung, J. Org. Chem. 2010, 75, 1281–1284; b) K. Ota,
S. I. Lee, J. M. Tang, M. Takachi, H. Nakai, T. Morimoto, H. Sakurai, K. Ka-
taoka, N. Chatani, J. Am. Chem. Soc. 2009, 131, 15203–15211.
[11] a) S. H. Sim, Y. J. Park, Y. K. Chung, Synlett 2012, 23, 473–477; b) Y. J.
Park, S. Y. Kim, J. E. Cho, J. H. Park, Y. K. Kang, Y. K. Chung, Chem.
Commun. 2011, 47, 5190–5192; c) S. Y. Kim, Y. K. Kang, Y. K. Chung,
Chem. Eur. J. 2010, 16, 5310–5313; d) S. M. Kim, J. H. Park, S. Y. Choi,
Y. K. Chung, Angew. Chem. 2007, 119, 6284–6287; Angew. Chem. Int. Ed.
2007, 46, 6172–6175; e) Y. J. Song, I. G. Jung, H. R. Lee, Y. T. Lee, Y. K.
Chung, H. Y. Jang, Tetrahedron Lett. 2007, 48, 6142–6146; f) S. I. Lee,
S. M. Kim, S. Y. Kim, M. R. Choi, Y. K. Chung, J. Org. Chem. 2006, 71,
9366–9372; g) S. M. Kim, S. I. Lee, Y. K. Chung, Org. Lett. 2006, 8, 5425–
5427; h) S. I. Lee, S. M. Kim, S. Y. Kim, Y. K. Chung, Synlett 2006, 2256–
2260; i) S. I. Lee, S. H. Sim, S. M. Kim, K. Kim, Y. K. Chung, J. Org. Chem.
2006, 71, 7120–7123; j) S. Y. Kim, Y. K. Chung, Angew. Chem. 2008, 120,
4992–4995; Angew. Chem. Int. Ed. 2008, 47, 4914–4917.
[31] a) Z. X. Yu, P. A. Wender, K. N. Houk, J. Am. Chem. Soc. 2004, 126, 9154;
b) X. Hong, P. Liu, K. N. Houk, J. Am. Chem. Soc. 2013, 135, 1456; c) X.
Hong, B. M. Trost, K. N. Houk, J. Am. Chem. Soc. 2013, 135, 6588.
[12] a) S. Y. Kim, Y. Park, S. Son, Y. K. Chung, Adv. Synth. Catal. 2012, 354,
179–186; b) S. Y. Kim, Y. Park, Y. K. Chung, Angew. Chem. 2010, 122,
425–428; Angew. Chem. Int. Ed. 2010, 49, 415–418.
[13] When the Pd/maleimide ratio was 1:1, 1:2, 1:3, 1:4, or 1:5 in the reaction
of 1a in toluene at 908C for 15 h, the yields of 1c were 10, 35, 35, 51,
and 46%, respectively.
Received: February 9, 2014
Published online on && &&, 2014
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
13
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
&
These are not the final page numbers! ÞÞ

Full Paper
FULL PAPER
Pd-catalyzed cycloisomerization of 1,6-
enynes has long been studied, which
provides a variety of 5-membered ring
systems. A method that transforms 1,6-
enynes to bicyclo[4.1.0]heptene-type
products through palladium catalysis
was elusive but eventually has been
found. A broad range of 4-aza-1,6-
enynes were successfully transformed to
3-aza-bicyclo[4.1.0]heptenes in moder-
ate-to-high yields when [PdCl2]/P(OPh)3
or [Pd(maleimidate)2(PPh3)2] catalysts
were used in a nonpolar solvent (see
figure).
&
Palladium
H. M. Oh, J. E. Park, J. Kim, J. H. Kim,
Y. K. Kang,* Y. K. Chung*
&& – &&
Pd-Catalyzed Cycloisomerization of 4-
Aza-1,6-enynes to 3-Aza-
bicyclo[4.1.0]hept-2-enes
&
&
Chem. Eur. J. 2014, 20, 1 – 14
www.chemeurj.org
14
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!