Intramolecular [2+2+2] cycloaddition reactions of Yne-ene-yne and Yne-yne-ene enediynes catalysed by RhI: Experimental and theoretical mechanistic studies

Article abstract of DOI:10.1002/chem.201102210

N-Tosyl-linked open-chain yne-ene-yne enediynes 1 and 2 and yne-yne-ene enediynes 3 and 4 have been satisfactorily synthesised. The [2+2+2] cycloaddition process catalysed by the Wilkinson catalyst [RhCl(PPh ₃)₃] was tested with the

Full text of DOI:10.1002/chem.201102210

FULL PAPER
DOI: 10.1002/chem.201102210
Intramolecular [2+2+2] Cycloaddition Reactions of Yne-ene-yne and Yne-
yne-ene Enediynes Catalysed by RhI: Experimental and Theoretical
Mechanistic Studies
Anna Dachs,[a, b] Anna Pla-Quintana,[a] Teodor Parella,[c] Miquel Solꢀ,*[a, b] and
Anna Roglans*[a]
In memory of Professor Rafael Suau
Abstract: N-Tosyl-linked open-chain
yne-ene-yne enediynes 1 and 2 and
yne-yne-ene enediynes 3 and 4 have
been satisfactorily synthesised. The
[2+2+2] cycloaddition process cata-
lysed by the Wilkinson catalyst [RhCl-
ACHTUNGTRENNUNG(PPh3)3] was tested with the above-
mentioned substrates resulting in the
production of high yields of the cyclo-
adducts. Enediynes 1 and 2 gave stan-
dard [2+2+2] cycloaddition reactions
whereas enediynes 3 and 4 suffered b-
hydride elimination followed by reduc-
tive elimination of the Wilkinson cata-
lyst to give cycloadducts, which are iso-
mers of those that would be obtained
by standard [2+2+2] cycloaddition re-
actions. The different reactivities of
these two types of enediyne have been
rationalised by density functional
theory calculations.
Keywords: cycloaddition · density
functional calculations · enediynes ·
reaction mechanisms · rhodium
Introduction
prevailed, leading to 1,3-cyclohexadienes of type A
(Scheme 1). However, when the alkene moiety was acyclic,
cyclohexadienes of type C were obtained. DFT calculations
The transition-metal-catalysed [2+2+2] cycloaddition reac-
tions of moieties consisting of two alkynes and an alkene
are a well-established method for the synthesis of 1,3-cyclo-
hexadienes. Several transition-metal complexes have been
reported as efficient catalysts in this process.[1–3] Among
them, rhodium catalysts are becoming increasingly popular[4]
especially in the enantioselective reaction in which [2+2+2]
cycloaddition reactions of diynes and alkenes,[5] enynes and
alkynes,[6] and enediynes[7] have been reported.
Recently, Saꢀ and co-workers studied the cycloaddition
reactions of a,w-diynes with alkenes by using a ruthenium
complex as catalyst to give cyclohexadiene products.[8] With
cyclic alkenes, the standard [2+2+2] cycloaddition pathway
Scheme 1. Two possible reaction pathways for the cycloaddition reactions
of diynes and alkenes
identified a common ruthenacycloheptadiene intermediate
(I) from which the two final products arose by either reduc-
tive elimination (product A) or b-hydride elimination fol-
lowed by a reductive elimination (product B). In the latter
case, the 1,3,5-triene (B) suffered a thermal disrotatory 6eꢀ
p electrocyclisation to afford cyclohexadiene derivative C.
In agreement with the discovery of this new reaction
pathway, Aubert, Gandon and co-workers[9] reported the cy-
cloaddition reactions of diynes with enol ethers catalysed by
stoichiometric [CpCoL2] (Cp=cyclopentadienyl).[9a] DFT
calculations confirmed that the energy barriers of the
[2+2+2] reductive and b-hydride elimination pathways are
similar, in line with previous results obtained by the same
authors.[9b]
[a] Dr. A. Dachs, Dr. A. Pla-Quintana, Prof. M. Solꢁ, Prof. A. Roglans
Departament de Quꢂmica, Universitat de Girona
Campus de Montilivi, s/n. 17071 Girona (Spain)
E-mail: miquel.sola@udg.edu
anna.roglans@udg.edu
[b] Dr. A. Dachs, Prof. M. Solꢁ
Institut de Quꢂmica Computacional, Universitat de Girona
Campus de Montilivi, s/n. 17071 Girona (Spain)
[c] Dr. T. Parella
Servei de RMN, Universitat Autꢃnoma de Barcelona
08193 Cerdanyola, Barcelona (Spain)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201102210. It contains NMR
spectra for compounds 1–4, 10, 11, 14 and 15, cartesian xyz coordi-
nates and total energies for all stationary points located. It also con-
tains Figures 4–13 in colour.
Chem. Eur. J. 2011, 17, 14493 – 14507
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
14493

In the case of rhodium complexes, Tanaka and co-workers
observed the formation of a hepta-2,4,6-trienamide in a
single case described in a footnote when they combined an
alkenylamide with a 1,6-diyne in the presence of [Rh-
two Boc groups, gave derivative 2 in a yield of 98%. Trisul-
fonamide 8 was treated with the allylic bromide 9[12b] in
K2CO3 and heated at reflux in acetonitrile to give derivative
3 in a yield of 75%. Elimination of the only Boc group led
to an almost quantitative yield of 4.
ACHTUNGTRENNUNG(cod)2]BF4/BIPHEP (BIPHEP=2,2ꢅ-bis(diphenylphosphi-
no)-1,1ꢅ-biphenyl).[10] In addition, Ojima and co-workers[11]
reported the formation of fused tetracyclic compounds start-
ing from enediynes and carbon monoxide by a RhI-catalysed
[2+2+2+1] cycloaddition reaction, which gave a 1:1 mixture
of regioisomers of the expected product and its diene-shifted
regioisomers. The authors did not give any mechanistic ex-
planation for the formation of the latter.
The aim of this work was to study the [2+2+2] cycloaddi-
tion reactions of a set of enediynes of type yne-ene-yne and
yne-yne-ene (see below) catalysed by RhI (in particular with
Once the corresponding unsaturated substrates had been
obtained, we studied their [2+2+2] cycloaddition reactions
(Scheme 3 and Table 1). The [RhClACHTNUTRGEN(UNG PPh3)3] complex was
used as it is simple, relatively inexpensive and commercially
available.
When catalytic amounts of [RhClACHTNUTRGEN(UNG PPh3)3] (10% molar)
were used in toluene at 1008C under anhydrous conditions,
the expected compounds 10 and 11 were obtained in good
yields (entries 1 and 2, Table 1). The simplicity of the 1H
and 13C NMR spectra proved a symmetrical structure due to
the presence of a C2 axis. The anti disposition of the new
the Wilkinson complex [RhClACHTUNTRGNEUNG(PPh3)3]) and to analyse
ꢀ
whether the reaction products are those that would be ex-
pected from a characteristic [2+2+2] cycloaddition reaction
or are derived from atypical b-hydride elimination pathways
leading to unconventional cyclohexadiene derivatives. To
understand the experimental results obtained, theoretical
calculations by using density functional theory (DFT) with a
hybrid functional were performed.
CHc CHc’ centres was established by taking into account
the trans configuration of the original double bond in the
starting compounds 1 and 2, respectively, and the NOE
data. The full characterisation of compounds 10 and 11 by
2D correlation NMR spectroscopy (see the Supporting In-
formation) confirmed that the process proceeded as a stan-
dard [2+2+2] cycloaddition reaction.
COSY data for 11 (see the experimental H and 13C NMR
1
chemical shifts for the ring moiety in Figure 1) confirmed
that the hydrogen atoms of the methylene groups 1 and 12
of the open chains (d=3.37 and 3.68 ppm) only couple with
the amine NH proton (d=7.74 ppm). Furthermore, the Hc/
Hc’ atoms of the cyclohexadiene ring are not coupled to
these methylenes, as was confirmed by HMBC data. NOE
data were essential to determine the relative anti stereo-
chemistry of the symmetrical centres in 10 and 11, in which
the Hc proton in compound 11 shows a substantial NOE
effect on the two closely situated Ha and Hb’ protons. This
behaviour confirmed the anti stereochemistry of the com-
pounds as the syn structure for compounds 10 and 11 would
have given a minimum NOE signal between the Hc and Ha/
Hb’ protons.
Results and Discussion
Synthesis: First, we applied our experience in the allylation
of sulfonamides to create macrocycles for the preparation of
1 and 3 and their N-Boc-protected derivatives 2 and 4.
Scheme 2 shows the synthetic approaches to both types of
enediyne (yne-ene-yne types 1 and 2 and yne-yne-ene types
3 and 4). In the two cases, the starting product was the bro-
mopropargylic derivative 5. We had previously prepared
compounds 5,[12a] 6,[12b] 7[12a] and 8.[12a] The treatment of 6
with an excess of bromide 5 in the presence of potassium
carbonate as base in acetonitrile at reflux resulted in ene-
diyne 1 in a yield of 75%, which, with the elimination of the
The cycloaddition reactions of azaenediynes 3 and 4 (en-
tries 3 and 4 in Table 1) were carried out in toluene at 858C
under anhydrous conditions. In these cases, complete analy-
sis of the NMR spectra confirmed that the expected deriva-
tives 12 and 13, respectively, were not formed. The products
resulting from these transformations were characterised as
their isomers, compounds 14 and 15, respectively (see
Figure 2 for the complete assignment of 15).
Abstract in Catalan: Sꢀhan preparat satisfactꢁriament una
sꢂrie dꢀendiins de cadena oberta amb unitats N-tosil i amb di-
ferents posicions dels triples enllaÅos en la cadena (1 i 2, amb
els triples enllaÅos no consecutius; 3 i 4, amb els triples enlla-
Åos consecutius). Aquests endiins sꢀhan emprat com a subs-
trats en processos de cicloaddiciꢃ [2+2+2] catalitzats pel
If the standard [2+2+2] cycloadduct 13 (R=H) had been
complex de Wilkinson [RhClACHTNUGTRENUNG(PPh3)3] obtenint-se bons rendi-
1
formed, the H and 13C NMR spectra would have been ex-
ments dels corresponents cicloadductes. Els endiins 1 i 2
donen una reacciꢃ de cicloaddiciꢃ [2+2+2] estꢄndard mentre
que els endiins 3 i 4 pateixen una b-eliminaciꢃ dꢀhidrur segui-
da dꢀuna eliminaciꢃ reductora del catalitzador de Wilkinson
generant els cicloadductes 14 i 15, els quals sꢃn isꢁmers dels
que sꢀobtindrien per una cicloaddiciꢃ [2+2+2] estꢄndard. La
diferent reactivitat dꢀaquest dos tipus dꢀendiins sꢀha raciona-
litzat mitjanÅant cꢄlculs teꢁrics basats en la teoria del funcio-
nal de la densitat.
pected to show certain features relating to compound sym-
metry. However, analysis and complete chemical shift as-
signment by 2D NMR techniques (see the Supporting Infor-
mation) provided evidence of a nonsymmetric compound.
Thus, COSY data confirmed that the different methylene
groups 1 and 12 are only coupled to the amine NH protons
(d=7.37 and 7.58 ppm, respectively). A key feature of the
results was the HMBC three-bond crosspeaks between the
C12 and C10 positions. Moreover, two different protons Hc
14494
www.chemeurj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14493 – 14507

Intramolecular Cycloadditions of Enediynes
FULL PAPER
Scheme 2. Synthesis of enediynes 1–4 (TFA=trifluoroacetic acid; Ts=tosyl; Boc=tert-butyloxycarbonyl).
spin system. NOE data were
used to confirm the stereo-
chemical assignments.
Computational
Given the experimental results
obtained, we performed
B3LYP/cc-pVDZ-PP calcula-
calculations:
tions (see the Computational
Methods section for a more de-
tailed description of the method
Figure 1. Selected
1H
and
13C NMR chemical shifts (the
latter in brackets) of 11 [ppm].
Scheme 3. RhI-catalysed cycloaddition reactions of compounds 1–4 to
yield 10, 11, 14 and 15.
Figure 2. Selected 1H and 13C NMR chemical shifts (the latter in brack-
ets) of 15 [ppm].
used) to unravel the reaction mechanism of the intramolecu-
lar [2+2+2] cycloaddition reactions of the enediynes 1–4
catalysed by the Wilkinson complex (Scheme 3) to under-
stand their different reactivities. To reduce the computation-
al effort required, the tosyl moieties present in the experi-
mental enediynes and the three phenyl groups of the Wil-
Table 1. RhI-catalysed cycloaddition of compounds 1–4 to yield 10, 11,
14, and 15.
Entry
Enediyne
T [8C]
Reaction time [h]
Product (yield [%])
1
2
3
4
1
2
3
4
100
100
85
48
48
30
28
10 (55)
11 (85)
14 (65)
15 (89)
kinson catalyst were substituted by hydrogen atoms.[13,14]
A
85
previous study[15] with macrocyclic systems revealed that the
substitution of the tosyl groups by hydrogen atoms reduces
the exothermicity of the [2+2+2] cycloaddition by about
10%. Although this quantity is not negligible, we expect it
to have only a slight effect on the different reaction mecha-
and Hd (d=2.73 and 1.86 ppm, respectively) are coupled to
each other and also to the corresponding vicinal CH2 groups
ꢀ
ꢀ
ꢀ
thereby confirming the presence of a CH2 CH CH CH2
Chem. Eur. J. 2011, 17, 14493 – 14507
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14495

M. Solꢁ, A. Roglans et al.
nisms studied here and therefore the conclusions reached by
our model systems will still be valid for experimental sys-
tems. Scheme 4 shows the model reactions studied in our
theoretical calculations.
catalyse this oxidative coupling step, we have considered
both possibilities.
ꢀ
The three C C oxidative couplings were analysed with
the [RhCl
A
couplings correspond to the coordination of alkene and
alkyne moieties (2-B1 and 2-B2, Figure 3), yielding the dis-
torted rhodacyclopentene adducts 2-C1 and 2-C2 via the
transition states (TSs) 2-TSACHTNUGRTNEUNG
(B1,C1) (DG° =32.9 kcalmolꢀ1
with respect to 2-A+[RhClACHTUNRGTENNUG(PH3)3]) and 2-TSACHTUNGTRENNUNG(B2,C2)
(DG° =34.9 kcalmolꢀ1), respectively. The third path, involv-
ing alkyne–alkyne coupling, transformed 2-B3 into 2-C3 via
2-TSACHTNUGRTNEUNG
(B3,C3) with a barrier of DG° =30.3 kcalmolꢀ1. The
small differences between the three Gibbs energy barriers
indicate a competition between the three couplings, with the
alkyne–alkyne coupling favoured, although only by about
3 kcalmolꢀ1.
The same three oxidative couplings were then studied
with [RhClACHTNUTRGNENG(U PH3)] as the catalytic species, that is, the species
in which the rhodium retains only one PH3 ligand
(Figure 4). Enyne couplings with complexes 2-B4 and 2-B5
led to the rhodacyclopentene adducts 2-C4 and 2-C5 via 2-
TSACHTNUTRGENNUG HCATUNGTRENNGUN
(B4,C4) (DG° =30.1 kcalmolꢀ1) and 2-TS(B5,C5) (DG° =
38.0 kcalmolꢀ1), respectively. On the other hand, the
Scheme 4. Reaction models 2-A and 4-A for the RhI-catalysed cycloaddi-
tion reactions of compounds 1–4.
alkyne–alkyne coupling step transformed 2-B into 2-C via 2-
TSACTHNUTRGNEUNG
(B,C) with DG° =28.0 kcalmolꢀ1.
Note that the preferred pathway is the alkyne–alkyne cou-
pling when the rhodium has either one or two phosphine li-
gands attached, which have barriers of 28.0 and 30.3 kcal
molꢀ1, respectively. The small energy difference seems to in-
Yne-ene-yne enediynes 1 and 2: We first examined the reac-
tion mechanism for enediynes 1 and 2, which were found to
react by a standard [2+2+2] cycloaddition process. These
two compounds differ in the protecting Boc groups on the
terminal amines connected to the alkyne moieties, so we
studied a model of these enediynes with terminal amines,
model 2-A in Scheme 4.
dicate that both the reaction mechanisms via 2-TS
with the [RhCl(PH3)] catalytic species, and 2-TS(B3,C3),
with the [RhCl(PH3)2] catalyst, may be operative. Note,
however, that the oxidative addition is both kinetically and
ACHTUNGTRENNUNG(B,C),
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
thermodynamically favoured with the [RhClACHTUNRGTNEUNG(PH3)] catalytic
There are two possibilities for these enediynes in the ini-
species. In a previous study,[16] we analysed the oxidative
ꢀ
tial C C oxidative coupling step: either alkyne–alkyne cou-
coupling step for the same 2-A system but with hydrogen
ꢀ
pling between two distant alkyne moieties or alkyne–alkene
(enyne) coupling due to the coordination of the alkene and
one alkyne moiety. In the latter, the alkene moiety has two
different faces that can be coordinated (see Figure 3), so
there are two possible enyne couplings.[16] The catalysis pro-
ceeds only if the Wilkinson catalyst loses one or two of the
three initially coordinated phosphine ligands. Because both
and CH3 groups instead of the CH2 NH2 groups as substitu-
ents on the alkyne moieties. For terminal alkynes (H instead
ꢀ
of CH2 NH2), the barriers to the alkyne–alkyne coupling
were 31.9 and 25.4 kcalmolꢀ1 for the [RhCl
(PH3)2] and
[RhCl
alkynes with methyl groups, the barriers were higher; the
[RhCl(PH3)2] catalyst favoured the enyne coupling with
DG° =36.4 kcalmolꢀ1, whereas the alkyne–alkyne coupling
ACHTUNGTREN(NUNG PH3)] catalysts, respectively, whereas in the case of
AHCTUNGTRENNUNG
the resulting active species, [RhClACTHNUTRGNEN(UG PH3)2] or [RhClAHCTUNGTREN(NUGN PH3)],
was preferred with the [RhCl
G
ꢀ1
32.6 kcalmol . On the whole, with H, CH3 and CH2 NH2 in
the enediynes of type 2-A, the alkyne–alkyne coupling with
the lowest barrier corresponds to the terminal alkynes with
the [RhClACTHNUTRGNEUNG
(PH3)] catalyst (DG° =25.4 kcalmolꢀ1) and to the
ꢀ
CH2 NH2 substituent with the [RhCl
N
ACHTUNERGN(UNG PPh3)3] and [RhClHCAUTGNTREN(NUNG PH3)3]. The main difference we found in
the Gibbs energy profiles for the two catalysts is the initial
phosphine dissociation step,[17] which is very favoured with
Figure 3. Schematic representation for the alkyne–alkene coordination
(2-B1 and 2-B2) of rhodium with the two faces of the alkene moiety.
the [RhClACTHNGUTERNNUG
(PPh3)3] catalyst.[13,14,17] This result is not unexpect-
14496
www.chemeurj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14493 – 14507

Intramolecular Cycloadditions of Enediynes
FULL PAPER
Figure 4. Gibbs energy profiles for the different oxidative coupling steps: Enyne couplings via 2-TSACTHUNRTGENNUG(B1,C1) and 2-TSACHUTNGTREN(NGUN B2,C2) when the rhodium catalyst
has two attached phosphine ligands, and via 2-TS
(B3,C3) when the rhodium catalyst has two phosphine ligands and via 2-TS
distances in ꢆ
ACHUTGTNREN(UNG B4,C4) and 2-TSACHTNUGTRENNUNG
N
ACHTUNGTRENNUNG
ed given the different electronic and steric effects exerted
by the PPh3 ligands as compared with PH3. Our previous
study[13] led us to expect that the Gibbs energy profiles for
the [2+2+2] cycloaddition reac-
energy barrier of DG° =15.5 kcalmolꢀ1 to give the rhodacy-
cloheptadiene intermediate 2-E via 2-TS(D,E), leaving only
one face to be coordinated due to system constraints.
AHCTUNGTRENNUNG
tions of the enediynes studied
would be slightly lower if com-
puted with the [RhCl
ACHTUNGTRNE(NUNG PPh3)3]
catalyst.[13,14,17]
If we study the 16-electron 2-
C
complex (arising from
alkyne–alkyne oxidative addi-
tion with [RhCl(PH3)]), two
ACHTUNGTRENNUNG
possibilities emerge from this
point (Figures 5 and 6). First, a
phosphine ligand can be at-
tached to complex 2-C to yield
2-C3. This process is endoer-
gonic by 1.2 kcalmolꢀ1. Second-
ly, intramolecular coordination
of the alkene moiety to yield
complex 2-D can take place
with an energy stabilisation of
26.6 kcalmolꢀ1. Therefore, coor-
dination of the alkene is more
favoured than the attachment
of an incoming PH3 ligand. The
Figure 5. Gibbs energy profiles (in kcalmolꢀ1) for the two possible insertions of an alkene moiety: From 2-D
to form complex 2-E via the transition state 2-TS(D,E) and from complex 2-D3 to form 2-E3 via 2-TS(D3,E3)
N
ACHTUNGTRENNUNG
olefin is then inserted with an for enediyne model 2-A.
Chem. Eur. J. 2011, 17, 14493 – 14507
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14497

M. Solꢁ, A. Roglans et al.
Figure 6. Structures of the reactants, intermediates, transition states and products optimised at the B3LYP level of theory for the transformation of 2-C
to 2-E and of 2-C3 to 2-E3 with selected bond distances [ꢆ].
On the other hand, if we study the 18-electron complex 2-
C3 (arising from alkyne–alkyne oxidative addition and
take place in this case. Starting from complex 2-E3, ring clo-
sure takes place via 2-TS(E3,F3) with a Gibbs energy barri-
AHCTUNGTRENNUNG
[RhCl
N
er of 24.9 kcalmolꢀ1 (Figure 7), although it is an exoergonic
step (DG8=ꢀ12.1 kcalmolꢀ1). This process yields complex
2-F3 in which the rhodium atom is coordinated through a
three-centre–four-electron (3c–4e) intramolecular metal–hy-
drogen bond (IMHB) to the Ha atom of the cyclohexadiene
alkene moiety to yield complex 2-D3 releases 20.8 kcal
molꢀ1. The olefin is then inserted into complex 2-D3 to
afford the rhodacycloheptadiene complex 2-E3 via 2-TS-
ACHTUNGTRENNUNG
(D3,E3) with a barrier of DG° =8.1 kcalmolꢀ1. With these
ꢀ
ꢀ
results we observe that the two reaction pathways have
slightly different energy requirements but that they could
both be operative.
Continuing the DFT calculations from the complexes 2-E
and 2-E3, the next step is ring closure (Figure 7). At this
point we observe that 2-E and 2-E3 have no hydrogen
atoms at the b position that can be eliminated (the closest
Hb is at 3.38 and 3.48 ꢆ, respectively, from the rhodium;
Figure 6). For this reason, b-hydride elimination could not
ring formed (the values of Rh Ha 1.86 ꢆ, Rh C 2.88 ꢆ and
aCHaRh 143.58 rule out the possibility of an agostic inter-
action and are typical of IMHBs[18]). Finally, the catalytic
cycle is closed upon exoergonic displacement (by 27.3 kcal
molꢀ1) of the cyclohexadiene product 2-H by a phosphine
molecule to regenerate the [RhCl
When the reaction mechanism evolves from intermediate
2-E via 2-TS
(E,F) (DG° =13.1 kcalmolꢀ1) a fused tricyclic
complex 2-F, in which the arene ring is coordinated in an h4
ACHTUGNTRENUN(GN PH3)3] catalyst (Figure 7).
AHCTUNGTRENNUNG
14498
www.chemeurj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14493 – 14507

Intramolecular Cycloadditions of Enediynes
FULL PAPER
have a significant influence on
the reaction mechanism as the
two enediynes 3 and 4 behave
similarly in the presence of the
Wilkinson catalyst.
As for model 2-A, we ana-
lysed the different possibilities
for the oxidative coupling step
in model 4-A (Scheme 4). Thus,
as can be seen in Figure 9, the
Wilkinson catalyst can interact
with the two adjacent triple
bonds or with the double bond
and the consecutive triple bond
of the enediyne. In addition, we
have taken into account the
fact that both [RhCl
ACHTUGNERTN(NUNG PH3)2] and
[RhCl(PH3)] may be the active
AHCTUNGTRENNUNG
species of the catalyst, as we
have seen previously.
When the catalytic species is
[RhClACHTUNTRGNEUNG(PH3)2], the most favour-
Figure 7. Gibbs energy profiles (in kcalmolꢀ1) for the formation of the product 2-H from 2-E and 2-E3.
ꢀ
able of the three C C oxidative
coupling reactions is the
fashion, is obtained. Complex 2-F then evolves to the cyclo-
hexadiene product (2-H) by adding two phosphine mole-
alkyne–alkyne coupling that transforms 4-B into 4-C via 4-
TS
(B,C) with the lowest barrier of DG° =22.1 kcalmolꢀ1. In
this process the [RhCl(PH3)3] catalyst first loses a PH3
AHCTUNGTRENNUNG
cules to regenerate the [RhCl
G
ACHTUNGTRENNUNG
termediates 2-F and 2-F3 are different complexes although
both transition states correspond to ring closure. Therefore
the pathway followed depends on the number of PH3 li-
gands attached to the rhodium atom.
The complete reaction mechanism for the rhodium-cata-
lysed reaction of enediynes 1
ligand and then interacts with 4-A to yield a distorted trigo-
nal-bipyramidal complex 4-B by replacing one phosphine
with two internal h2 interactions with adjacent acetylenic
units of 4-A. Note that the transformation of 4-A into 4-B is
endoergonic by 16.7 kcalmolꢀ1, although this endoergonicity
and 2 deriving from our analy-
sis is presented in Figure 8.
There are two main pathways
that yield 2-H from 2-A that
differ in the active catalyst:
[RhCl
[RhClACHTUNGTRENNUNG
ACHTUNGTRENNUNG
results show that both pathways
may be operative given the sim-
ilar energy requirements.
Yne-yne-ene enediynes 3 and 4:
We then examined the reaction
mechanism for enediynes 3 and
4, which did not afford the stan-
dard [2+2+2] cycloadducts.
These two compounds only
differ in the protecting group
(Boc) in the terminal amine
connected to the alkene. We
studied a model of these ene-
diynes with primary amines
(model 4-A in Scheme 4). The
Figure 8. Gibbs energy profiles for the [2+2+2] cycloaddition of model 2-A (enediynes 1 and 2) in which the
active catalyst is [RhCl
ACHTUTGNREN(NUG PH3)2] (dashed line) or [RhClACHTUNGTRENN(UGN PH3)] (solid line). Relative Gibbs energies with respect
to separated reactants in kcalmolꢀ1. Imaginary frequencies [cmꢀ1] for the different transition states are given
Boc group does not seem to in brackets.
Chem. Eur. J. 2011, 17, 14493 – 14507
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14499

M. Solꢁ, A. Roglans et al.
from species 4-C given that our
previous study of model 2-A re-
vealed that the reaction mecha-
nism is similar in both cases.
The oxidative coupling of the
two alkynes 4-B!4-C leads to
the distorted trigonal-bipyrami-
dal rhodacyclopentadiene 4-C
with
a
Gibbs stabilisation
energy of 34.9 kcalmolꢀ1. In the
next step, the formation of two
stereoisomers (4-D and 4-D1) is
possible through the coordina-
tion of the two faces of the
olefin, which now have no
steric constraints. Moreover,
the olefin can be inserted into
each one of the two nonequiva-
ꢀ
lent Rh C bonds of the rhoda-
cyclopentadiene.
For
this
reason, the energies of the
alkene–rhodacyclopentadiene
stereoisomers 4-D and 4-D1
were computed with the corre-
Figure 9. Gibbs energy profiles (in kcalmolꢀ1) for the different oxidative coupling steps: Enyne couplings via
4-TS(B1,C1) and 4-TS(B2,C2) when the rhodium has two phosphine ligands attached and via 4-TS(B4,C4)
and 4-TS(B5,C5) when there is only one PH3 ligand attached, and alkyne–alkyne couplings via 4-TS(B,C)
(B3,C3) when the rhodium has only one phosphine
G
R
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
ꢀ
sponding TSs of the C C cou-
when the rhodium has two phosphine ligands and via 4-TSACHTUNGTRENNUNG
attached. Energies in kcalmolꢀ1 and distances in ꢆ.
pling reactions leading to the
fused bicyclic ring systems 4-E,
4-E1,
4-E2
and
4-E3
is probably overestimated as a result of substitution of the
PPh3 ligands in the Wilkinson catalyst by a stronger s
donor, such as PH3.[17] The other two possible oxidative ad-
dition reactions correspond to the coordination of the
alkene and alkyne moieties (4-B1 and 4-B2) to yield the dis-
torted rhodacyclopentene adducts 4-C1 and 4-C2 via 4-TS-
(Scheme 5). The [4+2] cycloaddition reaction was also in-
vestigated, but all attempts were unsuccessful and no TS for
this route was located. We therefore discarded this possible
pathway and focused on the pathways shown in Scheme 5.
The rhodacyclopentadiene complexes 4-D and 4-D1 have
similar stabilities with respect to the reactants. Of the differ-
ent activation processes, the one with the greatest cost in
terms of energy is the transformation of 4-D1 into 4-E2 via
A
(B2,C2) (DG° =
ACHTUNGTRENNUNG
4-TS
ACHTUNGTRENNUNG
A
formation of 4-D1 into 4-E3 via 4-TSACTHUNGTRENNNUG
16.5 kcalmolꢀ1). We were unable to locate the TS corre-
sponding to the transformation of 4-D into 4-E1 and all our
attempts led to the TS for the transformation of 4-D into 4-
E, which has the lowest energy barrier of DG° =8.1 kcal
molꢀ1. Therefore, we concluded that the preferred pathway
ACHTUNGTRENNUNG
most favourable oxidative coupling. The enyne couplings of
complexes 4-B4 and 4-B5 leading to the rhodacyclopentene
adducts 4-C4 and 4-C5 via 4-TS
(B4,C4) (DG° =35.4 kcal
ACHTUNGTRENNUNG
molꢀ1) and 4-TS(B5,C5) (DG° =28.8 kcalmolꢀ1), respective-
G
is the C C coupling reaction between the alkene moiety co-
ꢀ
ly, have energy barriers that are higher by about 10–15 kcal
molꢀ1.
ordinated to the rhodacyclopentadiene 4-D via 4-TS
which leads to the rhodabicyclic ring system 4-E.
ACHTUNGTRENNUNG(D,E),
There is a clear preference for the alkyne–alkyne coupling
over the enyne coupling reactions irrespective of whether
the [RhClACHTUNGTRENNUNG(PH3)2] or [RhClACHUTNGTREN(NUGN PH3)] catalytic species is used, in
There are two different pathways from complex 4-E
ꢀ
ꢀ
(Figure 10). The first is a C C bond formation and a Rh C
bond cleavage assisted by an Hb agostic interaction via the
agreement with a previous study showing that alkyne–
alkyne coupling is usually preferred.[16] Our results therefore
show that the pathway following the enyne coupling can be
ruled out and for this reason we continued our study from
species 4-C, which is the most stable species resulting from
the different oxidative couplings analysed. Although species
4-C3 is kinetically favoured (although not thermodynamical-
ly) over 4-C, we have decided only to follow the analysis
transition state 4-TSACTHNGUTERNNUG
(E,F) with DG° =14.5 kcalmolꢀ1. This
process leads to 4-F in which the rhodium atom has a dis-
torted octahedral geometry and maintains an agostic inter-
ꢀ
ꢀ
action with Hb (the values of Rh Hb 1.88 ꢆ, Rh C 2.31 ꢆ
and aCHbRh 95.28 are typical of agostic interactions[18]).
This Hb agostic interaction is similar to that found for the
ruthenium system of Saꢀ and co-workers,[8b] which has simi-
ꢀ
ꢀ
lar distances (Ru Hb 1.82 ꢆ, Ru C 2.40 ꢆ and aCHbRu
14500
www.chemeurj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14493 – 14507

Intramolecular Cycloadditions of Enediynes
FULL PAPER
clic opening of the complex 4-E to give the rhodacyclohep-
tadiene 4-F1. The barrier of 4-TS
(E,F1) (DG° =7.2 kcal
ACHTUNGTRENNUNG
molꢀ1) is lower than that of 4-TS
AHCTUNGTRENNUNG
sults reported by Saꢀ and co-workers, for which this path-
way was also found to have the lower Gibbs energy (DG° =
12.2 kcalmolꢀ1).[8b]
Therefore, we followed the reaction pathway from both
intermediates 4-F and 4-F1 (Figures 11 and 12). From 4-F1
the ring can be closed via 4-TSACTHNUTRGNEUGN(F1,I2), which has a high bar-
rier of 24.5 kcalmolꢀ1, to give the 3c–4e IMHB intermediate
4-I2 (dashed line in Figure 11). This transformation has a
high energy barrier and, taking into account the fact that 4-
TSACTHNUGRTENUNG(E,F1) has a low barrier, the process that connects 4-E
with 4-F1 may be reversible and may follow the alternative
reaction pathway (4-F1!4-E!4-F!4-G!4-H), which has
lower-energy requirements.
From 4-F, we found two different pathways (Figures 11
and 12). First, path A (dotted line, 4-F!4-J1) involves re-
ductive elimination, leading to the cyclohexadiene product
after ring slippage and regeneration of the Wilkinson cata-
lyst (see the optimised intermediates in Figure 12). The TS
(DG° =7.0 kcalmolꢀ1) of this process (4-TS
ACTHNUTRGNE(NUG F,I1)) requires
the loss of one phosphine ligand, leading to a fused tricyclic
complex 4-I1 in which the arene ring is coordinated in an h4
fashion to the metal (see Figure 12). This intermediate then
evolves to the cyclohexadiene product 4-J1 adding two
phosphine molecules to regenerate the [RhClACHTNUGTRNEG(UN PH3)3] cata-
Scheme 5. Possible stereoisomers resulting from the coordination of the
rhodacyclopentadiene complex to the double bond of the open chain and
lyst. This compound, 4-J1, is the expected product of the
[2+2+2] cycloaddition (12 and 13 in Scheme 3) but not the
experimentally obtained product (14 and 15 in Scheme 3).
Secondly, path B (solid line in Figure 11) entails a b-hydride
ꢀ
possible C C couplings in each case. Gibbs energy values given relative
to 4-A are in kcalmolꢀ1
.
103.18) but a higher energy barrier for the transformation
(DG° =21.8 kcalmolꢀ1). The second pathway is an electrocy-
elimination from 4-F via 4-TSACTHGNUTERNNUG
(F,G) (DG° =5.8 kcalmolꢀ1)
to afford the complex 4-G as a rhodium hydride distorted
octahedral complex in which
the rhodium is coordinated in
an h3 fashion to the six-mem-
bered ring (see Figure 12).
Next, the release of one PH3
ligand is required to give the
tetra-coordinated complex 4-H
(stabilised by 22.8 kcalmolꢀ1
with respect to 4-G). Although
we have not located the TS cor-
responding to the 4-G!4-H
process, we expect a small barri-
er for this exergonic loss of a
phosphine ligand. The inter-
mediate 4-H has an adequate
conformation for reductive
elimination
via
4-TSACHTUNGTRENNUNG(H,I)
(DG° =11.5 kcalmolꢀ1), leading
to the fused tricyclic complex 4-
I in which the rhodium is coor-
dinated in an h2 fashion to the
arene ring and the metal also
has an agostic interaction with
Figure 10. Gibbs energy profiles (in kcalmolꢀ1) for the formation of complexes 4-F and 4-F1 via the transition
ꢀ
ꢀ
states 4-TS
(E,F) and 4-TS
G
Hb (Rh Hb 1.79 ꢆ, Rh C
Chem. Eur. J. 2011, 17, 14493 – 14507
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14501

M. Solꢁ, A. Roglans et al.
dition reactions of 1-ene-vinyl-
cyclopropanes. Similar to our
proposed mechanism, this b-hy-
dride elimination also involves
ꢀ
an intermediate with a Rh H
bond.
Conclusion
The paths of the rhodium-cata-
lysed intramolecular [2+2+2]
cycloaddition reactions of ene-
diynes 1–4 to give the corre-
sponding cyclohexadienes vary
with the position of the alkene
moiety in the enediyne. For
enediyne model 2-A (yne-ene-
yne), the standard [2+2+2] cy-
cloaddition reaction giving
compounds 10 and 11 is pre-
ferred as the key intermediates
2-E and 2-E3 cannot undergo
b-hydride elimination due to
Figure 11. Gibbs energy profiles (in kcalmolꢀ1) for the formation of 4-J which is the model for products 14
and 15.
2.38 ꢆ and aCHbRh 103.28). The displacement of the arene
4-J and the regeneration of the [RhCl(PH3)3] catalyst is the
the large distance between the rhodium and Hb atoms. In
contrast, the enediyne model 4-A (yne-yne-ene) undergoes
b-hydride elimination followed by reductive elimination of
the Wilkinson catalyst to yield cycloadducts 14 and 15,
which are isomers of the products that would be obtained
by standard [2+2+2] cycloaddition reactions. In this case,
the key intermediate is the complex 4-F in which an agostic
interaction between rhodium and hydrogen allows b-hydride
elimination. Experimental observations and DFT calcula-
tions support the mechanism proposed.
AHCTUNGTRENNUNG
next conversion. This compound, 4-J, coincides with the
product obtained experimentally (14 and 15 in Scheme 3;
see the optimised intermediates in Figure 12), which is ther-
modynamically 0.6 kcalmolꢀ1 more stable than the expected
product (4-J1). Therefore, path B accounts for the rhodium-
catalysed reactions of enediynes 3 and 4 affording the final
experimental product observed. Both thermodynamically
and kinetically, 4-J is slightly favoured over 4-J1 by a few
kcalmolꢀ1. We also analysed the process that does not stop
at the rhodium hydride 4-H, in which Hb is transferred di-
rectly to the a carbon to give complex 4-I3 (direct transfor-
mation 4-F-4-I3 is not shown in Figure 11). Given that the
Experimental Section
energy barrier of this process is too high (4-TSACTHNUTRGNEUNG
(F,I3), DG° =
Unless otherwise noted, materials were obtained from commercial sup-
pliers and used without further purification. All reactions requiring anhy-
drous conditions were conducted in oven-dried glassware under a dry ni-
trogen atmosphere. All solvents were distilled over appropriate drying re-
agents (sodium or calcium hydride) in an inert atmosphere. The solvents
were removed under reduced pressure with a rotary evaporator. Residues
were purified by chromatography on a silica gel column (230–400 mesh)
by using a gradient solvent system (hexane/ethyl acetate or hexane/di-
chloromethane) as the eluent.
1H and 13C NMR spectra were measured on a 600 or a 200 MHz NMR
spectrometer. 1H and 13C chemical shifts (d) are referenced to internal
solvent resonances and reported relative to SiMe4. The chemical shifts
were assigned on the basis of 2D COSY, NOESY, HSQC and HMBC ex-
periments performed under routine conditions.
30.7 kcalmolꢀ1) for it to take place instead of the 4-F!4-
G!4-H!4-I pathway, we discarded the possibility of this
direct transformation being operative.
The energy profile for the whole mechanism of the reac-
tions of the enediynes 3 and 4 catalysed by [RhClACHTNUGTRNEG(UN PH3)3] is
presented in Figure 13 whereas Scheme 6 provides a summa-
ry of the catalytic cycle proposed for the [2+2+2] cycloaddi-
tion reactions of enediynes 3 and 4. There are several differ-
ences between the present b-elimination mechanism and
that found for Cp*Ru (Cp*=1,2,3,4,5-pentamethylcyclopen-
tadienyl) and CpCo systems. First, we did not observe met-
allacycloheptadiene but rather bicyclic intermediates. Sec-
ondly, our final product is the cyclohexadiene product,
whereas for the other systems the final product is a 1,3,5-
hexatriene species. Finally, it is pertinent to point out that a
recent B3LYP/LANL2DZ-6-31G* study[19] has shown that a
similar b-hydride elimination can compete with the [3+2]
pathway in the RhI-catalysed intramolecular [3+2] cycload-
N-(4-Bromo-2-butynyl)-N-(tert-butyloxycarbonyl)-4-methylphenylsulfon-
amide (5), 1,11-bis(tert-butyloxycarbonyl)-1,6,11-tris(4-methylphenylsul-
fonyl)-1,6,11-triazaundeca-3,8-diyne (7) and 1,6,11-tris(4-methylphenyl-
sulfonyl)-1,6,11-triazaundeca-3,8-diyne (8) were prepared as previously
reported by our group.[12a] (E)-N,N’-Bis(4-methylphenylsulfonyl)-2-
butene-1,4-diamine (6)[12b] and N-[(E)-4-bromo-2-butenyl]-N-(tert-butyl-
oxycarbonyl)-4-methylphenylsulfonamide (9)[12b] were prepared as previ-
ously reported.
14502
www.chemeurj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14493 – 14507

Intramolecular Cycloadditions of Enediynes
FULL PAPER
Figure 12. Structures of the intermediates optimised at the B3LYP level of theory for the transformation of 4-F1 into 4-I2 and the intermediates of
paths A and B with selected bond distances [ꢆ].
1,16-Bis((tert-butyloxycarbonyl)-1,6,11,16-tetrakis(4-methylphenylsulfon-
yl)-1,6,11,16-tetraazahexadeca-8-ene-3,13-diyne (1): A stirred mixture of
6 (0.60 g, 1.52 mmol), potassium carbonate (1.08 g, 7.61 mmol) and aceto-
nitrile (40 mL) was heated at reflux for 10 min. Then a solution of N-(4-
bromo-2-butynyl)-N-(tert-butyloxycarbonyl)-4-methylphenylsulfonamide
(5; 1.23 g, 3.05 mmol) in acetonitrile (10 mL) was added slowly to the re-
action mixture. The reaction was heated and monitored by TLC until
completion (20 h). The salts were filtered off and the filtrate was evapo-
rated. The residue was purified by column chromatography on silica gel
with hexanes/ethyl acetate (polarity from 8:2 to 6:4) to afford 1 (1.20 g,
75%) as a colourless solid. M.p. 78–808C; 1H NMR (200 MHz, CDCl3,
258C, TMS): d=1.32 (s, 18H), 2.42 (s, 6H), 2.44 (s, 6H), 3.70–3.79 (m,
4H), 4.05 (br s, 4H), 4.39 (br s, 4H), 5.55–5.61 (m, 2H), 7.27–7.35 (m,
8H), 7.71 (AA’BB’ system, 3J
ACHTUNGTRENNUNG
system, 3J
ACHTUNGTRENNUNG
d=22.2, 22.3, 28.5, 36.2, 37.0, 48.5, 77.4, 81.8, 85.7, 128.3, 128.6, 129.7,
130.1, 130.4, 136.6, 137.4, 144.5, 145.3, 150.8 ppm; IR (ATR): n˜ =2981,
1735, 1344, 1159 cmꢀ1; HRMS (ESI): m/z calcd for C50H60N4S4O12+Na+:
1059.2983; found: 1059.2947.
1,6,11,16-Tetrakis(4-methylphenylsulfonyl)-1,6,11,16-tetraazahexadeca-8-
ene-3,13-diyne (2): A mixture of 1 (0.80 g, 0.78 mmol), trifluoroacetic
acid (3.2 mL) and dichloromethane (10 mL) was stirred at room tempera-
Chem. Eur. J. 2011, 17, 14493 – 14507
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14503

M. Solꢁ, A. Roglans et al.
33.4, 37.2, 48.8, 78.2, 80.9, 127.9, 128.4,
129.8, 130.3, 130.4, 136.4, 137.2, 144.5,
144.7 ppm; IR (ATR): n˜ =3249, 2923,
1331, 1155 cmꢀ1
calcd for
;
HRMS (ESI): m/z
C40H44N4S4O8+Na+:
859.1934; found: 859.1893.
1,16-Bis(tert-butyloxycarbonyl)-
1,6,11,16-tetrakis(4-methylphenylsul-
fonyl)-1,6,11,16-tetraazahexadeca-3-
ene-8,13-diyne (3): A stirred mixture
of
8 (0.92 g, 1.50 mmol), potassium
carbonate (0.53 , 3.75 mmol) and ace-
tonitrile (100 mL) was heated at reflux
for 10 min. Then a solution of N-[(E)-
4-bromo-2-butenyl]-N-(tert-butyloxy-
carbonyl)-4-methylphenylsulfonamide
(9; 0.30 g, 0.75 mmol) in acetonitrile
(10 mL) was added slowly to the reac-
tion mixture. The reaction was heated
and monitored by TLC until comple-
tion (4 h). The salts were filtered off
and the filtrate was evaporated. The
residue was purified by column chro-
matography on silica gel with dichloro-
methane/ethyl acetate (polarity 40:1)
to afford 3 (0.50 g, 75%) as a colour-
less solid. M.p. 76–808C; 1H NMR
(200 MHz, CDCl3, 258C, TMS): d=
Figure 13. Gibbs energy profile for the [2+2+2] cycloaddition of the model of enediynes 3 and 4. Gibbs
energy values relative to the initial reactants are in kcalmolꢀ1. Imaginary frequencies [cmꢀ1] for the different
transition states are given in parentheses.
3
ture for 7 h (TLC monitoring). The liquid was distilled off under vacuum
1.32 (s, 9H), 2.44 (m, 12H), 3.58 (d, J-
3
and the residue was dissolved in ethyl acetate (20 mL). The organic layer
was subsequently washed with aqueous sodium bicarbonate (3ꢇ20 mL),
H2O (3ꢇ20 mL) and brine (20 mL), dried (Na2SO4) and filtered. The sol-
vent was evaporated under reduced pressure to afford 2 (0.60 g, 98%) as
a colourless solid. M.p. 85–888C; 1H NMR (200 MHz, CDCl3, 258C,
(H,H)=6 Hz, 2H), 3.70 (d, J
(H,H)=6 Hz, 2H), 3.79 (m, 4H), 3.95 (br s,
2H), 4.36 (d, 3J(H,H)=6 Hz, 2H), 4.73 (t, 3J
ACTHNUGTRENNUGN ACHTUNGTRENNNUG
2H), 7.32 (m, 8H), 7.60 (AA’BB’ system, 3J
G
3
3
(AA’BB’ system, J
(H,H)=8.4 Hz, 2H), 7.69 (AA’BB’ system, J
T
3
8.4 Hz, 2H), 7.74 ppm (AA’BB’ system, JACTHUNGTRENNNUG
TMS): d=2.42 (s, 6H), 2.44 (s, 6H), 3.52 (t, 3J
(t, 3J(H,H)=2 Hz, 2H), 3.60–3.69 (m, 4H), 3.87 (brs, 4H), 4.68 (t, 3J-
(H,H)=5.9 Hz, 2H), 5.50–5.61 (m, 2H), 7.27–7.35 (m, 8H), 7.66
(AA’BB’ system, 3J(H,H)=8.2 Hz, 4H), 7.71 ppm (AA’BB’ system, 3J-
(H,H)=8.4 Hz, 4H); 13C NMR (50 MHz, CDCl3, 258C, TMS): d=22.2,
ACHTUNGERTN(NUNG H,H)=1.8 Hz, 2H), 3.55
(50 MHz, CDCl3, 258C, TMS): d=20.7, 20.8, 27.1, 32.1, 35.5, 35.6, 46.9,
47.1, 76.1, 77.4, 77.9, 79.5, 83.9, 126.5, 126.8, 126.9, 127.0, 127.1, 128.6,
128.8, 128.9, 130.2, 134.3, 134.9, 135.9, 136.3, 143.0, 143.2, 143.5, 143.7,
149.9 ppm; IR (ATR): n˜ =3275, 2974, 1725, 1345, 1155 cmꢀ1; HRMS
(ESI): m/z calcd for C45H52N4S4O10+Na+: 959.2458; found: 959.2414.
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
Scheme 6. Catalytic cycle proposed for the intramolecular [2+2+2] cycloaddition reactions of enediynes 3 and 4.
14504
www.chemeurj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14493 – 14507

Intramolecular Cycloadditions of Enediynes
FULL PAPER
1,6,11,16-Tetrakis(4-methylphenylsulfonyl)-1,6,11,16-tetraazahexadeca-3-
ene-8,13-diyne (4): A mixture of 3 (0.30 g, 0.32 mmol), trifluoroacetic
acid (1.5 mL) and dichloromethane (4 mL) was stirred at room tempera-
ture for 3 h (TLC monitoring). The liquid was distilled off under vacuum
and the residue was dissolved in ethyl acetate (20 mL). The organic layer
was subsequently washed with aqueous sodium bicarbonate (3ꢇ20 mL),
H2O (3ꢇ20 mL) and brine (20 mL), dried (Na2SO4) and filtered. The sol-
vent was evaporated under reduced pressure to afford 4 (0.23 g, 96%) as
a colourless solid. M.p. 70–748C; 1H NMR (200 MHz, CDCl3, 258C,
127.5, 128.6, 129.4, 129.6, 129.7, 132.0, 132.2, 133.5, 136.3, 137.2, 142.7,
143.8, 144.7, 150.7, 166.9 ppm; IR (ATR): n˜ =3321, 2920, 1728, 1346,
1157 cmꢀ1; HRMS (ESI): m/z calcd for C45H52N4S4O10+Na+: 959.2458;
found: 959.2413.
Cyclohexadiene 15: Column chromatography: hexanes/dichloromethane/
ethyl acetate (5:3:2) to afford 15 (0.05 g, 89%) as a colourless solid. M.p.
107–1098C; 1H NMR (600 MHz, CDCl3, 258C, TMS): d=1.86 (d, 3J-
3
G
N
3H), 2.40 (s, 3H), 2.41 (s, 3H), 2.69–2.73 (m, 2H), 3.09 (m, 1H), 3.26
(dd, 3J(H,H)=5.0 Hz, 2J(H,H)=15.0 Hz, 1H), 3.48 (m, 1H), 3.55 (d, 2J-
(H,H)=15.5 Hz, 1H), 3.70–3.80 (m, 3H), 3.84 (t, J=8.8 Hz, 1H), 3.90–
3.93 (m, 2H) , 7.17 (d, 3J(H,H)=8.0 Hz, 2H), 7.37 (br s, 1H), 7.38 (d, 3J-
(H,H)=8.0 Hz, 2H), 7.41 (d, 3J(H,H)=8.0 Hz, 4H), 7.45 (d, 3J
(H,H)=
8.0 Hz, 2H), 7.58 (t, 3J
(H,H)=5.3 Hz, 1H), 7.67–7.70 ppm (m, 6H);
TMS): d=2.43 (s, 6H), 2.45 (s, 6H), 3.52 (m, 4H), 3.65 (d, 3J
A
A
ACHTUNGTRENNUNG
3
6 Hz, 2H), 3.82 (m, 4H), 3.91 (s, 2H), 4.89 (t, J
(H,H)=5.2 Hz, 2H), 5.55
(m, 2H), 7.31 (m, 8H), 7.62 (AA’BB’ system, JACTHNUTRGENUG(N H,H)=8.4 Hz, 2H), 7.66
AHCTUNGTRENNUNG
3
AHCTUNGTRENNUNG
3
3
(AA’BB’ system, J
ACHTUNGTRENNUNG(H,H)=8.4 Hz, 2H), 7.70 (AA’BB’ system, JACHTUNGTRENNUNG
A
R
ACHTUNGTRENNUNG
3
8.2 Hz, 2H), 7.72 ppm (AA’BB’ system, JAHCUTNGTRENNUNG
AHCTUNGTRENNUNG
13C NMR (150 MHz, CDCl3, 258C, TMS): d=21.6, 21.9, 21.9, 21.9, 40.8,
41.6, 41.8, 42.7, 49.6, 50.0, 50.7, 54.7, 123.6, 124.3, 127.2, 127.4, 128.5,
128.6, 128.7, 130.1, 130.5, 130.7, 130.8, 132.7, 133.3, 134.3, 138.3, 138.3,
143.5, 143.6, 144.6, 144.6 ppm; IR (ATR): n˜ =3296, 2926, 1333,
1157 cmꢀ1; HRMS (ESI): m/z calcd for C40H44N4S4O8+Na+: 859.1934;
found: 859.1934; m/z calcd for C40H44N4S4O8+K+: 875.1674; found:
875.1681.
(50 MHz, CDCl3, 258C, TMS): d=21.5, 32.8, 36.3, 36.4, 36.5, 44.4, 48.2,
126.9, 127.1, 127.2, 127.6, 127.9, 129.6, 129.7, 129.8, 130.9, 135.1, 135.7,
136.4, 136.7, 143.6, 143.9, 144.0, 144.2 ppm; IR (ATR): n˜ =3283, 2922,
1326, 1154 cmꢀ1
;
HRMS (ESI): m/z calcd for C40H44N4S4O8+Na+:
859.1934; found: 859.1897.
General method for the cycloaddition reactions of 1–4: A degassed solu-
tion of enediyne (0.05 mmol) and chlorotris(triphenylphosphane)
rhodium(I) (0.005 mmol, 10% molar) in anhydrous toluene (10 mL) was
heated (temperatures and reaction times specified in Table 1) until com-
pletion (TLC monitoring). The solvent was then evaporated and the resi-
due was purified by column chromatography on silica gel.
Computational methods: All geometry optimisations were performed by
using the hybrid DFT B3LYP[20] method with the Gaussian03[21] program
package. The geometry optimisations were performed without symmetry
constraints. Analytical Hessians were computed to determine the nature
of the stationary points (one or zero imaginary frequencies for transition
states and minima, respectively) and to calculate unscaled zero-point en-
ergies (ZPEs), as well as thermal corrections and entropy effects using
the standard statistical mechanics relationships for an ideal gas.[22] These
two latter terms were computed at 298.15 K and 1 atm to provide the rel-
ative Gibbs energies (DG298). Furthermore, the connectivity between sta-
tionary points was established by calculations of the intrinsic reaction
paths.[23] The all-electron cc-pVDZ basis set was used for phosphorus,
oxygen, nitrogen, carbon, and hydrogen atoms,[24] whereas for rhodium
we employed the cc-pVDZ-PP basis set[25] containing an effective core
relativistic pseudopotential. Relative energies were computed by taking
into account the total number of molecules present. The SO2Ar moieties
present in the experimental enediynes and the phenyl group in the cata-
lyst were substituted by hydrogen atoms to reduce the computational
complexity of the calculations involving these ligands. Substitution of
Cyclohexadiene 10: Column chromatography: from hexanes/dichlorome-
thane (7:3) to hexanes/dichloromethane/ethyl acetate (7:3:1) to afford 10
(0.03 g, 55%) as a colourless solid. M.p. 90–938C; 1H NMR (600 MHz,
[D6]DMSO, 258C): d=1.31 (s, 18H), 2.38 (s, 6H), 2.39 (s, 6H), 2.41 (m,
2H), 2.55 (m, 2H), 3.69 (d, 3J
ACTHUNTGRENNG(U H,H)=15 Hz, 2H), 3.77 (m, 2H), 4.00 (d,
3
3
3J
16.2 Hz, 2H), 7.38 (AA’BB’ system, J
system, 3J(H,H)=7.8 Hz, 4H), 7.63 (AA’BB’ system, 3J
4H), 7.66 ppm (AA’BB’ system, 3J(H,H)=8.4 Hz, 4H); 13C NMR
ACHTUNGTRENNUNG(H,H)=15 Hz, 2H), 4.43 (d, JACHTUNREGTG(NNUN H,H)=16.2 Hz, 2H), 4.50 (d, JACTUHNGTRENNUGN
3
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
(150 MHz, [D6]DMSO, 258C): d=21.0, 27.1, 42.7, 44.6, 49.1, 52.4, 125.4,
127.3, 127.6, 129.5, 129.8, 131.8, 136.0, 136.2, 143.7, 144.7, 150.3 ppm; IR
(ATR): n˜ =2921, 1727, 1347, 1153 cmꢀ1; HRMS (ESI): m/z calcd for
C50H60N4S4O12+Na+: 1059.2983; found: 1059.2944; m/z calcd for
C50H60N4S4O12+K+: 1075.2722; found: 1075.2685.
Cyclohexadiene 11: Column chromatography: from hexanes/dichlorome-
thane/ethyl acetate (5:3:1) to hexanes/dichloromethane/ethyl acetate
(7:3:2) to afford 11 (0.04 g, 85%) as a colourless solid. M.p. 106–1088C;
1H NMR (600 MHz, [D6]DMSO, 258C): d=1.35 (m, 2H), 2.11 (m, 2H),
PPh3 by PH3 is
a common procedure in theoretical organometallic
chemistry.[26,27] In addition, we have checked that, despite the electronic
and steric differences, substitution of PPh3 by PH3 does not introduce sig-
nificant changes in the thermodynamics and kinetics of the cycloaddition
of three acetylene molecules.[14] A previous study found that solvent ef-
fects due to toluene and acetonitrile in [2+2+2] cycloadditions are
minor, likely due to the absence of charged or polarised intermediates
and transition states in the reaction mechanism.[28] Because the reactions
studied were carried out in toluene, solvent effects have not been includ-
ed in the present calculations. Finally, because there are no experimental
data suggesting the presence of paramagnetic intermediates, our studies
were limited to the singlet potential energy surfaces.
2.27 (s, 6H), 2.44 (s, 6H), 3.37 (m, 2H), 3.40 (m, 2H), 3.60 (d, 3J
15 Hz, 2H), 3.68 (dd, 3J(H,H)=6 Hz, 2J
(H,H)=15 Hz, 2H), 7.07 (AA’BB’ system, 3J
ACHTUNGTRENNUNG
G
ACHTUNGTRENNUNG
E
ACHTUNGTRENNUNG
3
3
(AA’BB’ system, J
8.4 Hz, 4H), 7.67 (AA’BB’ system, J
(H,H)=6 Hz, 2H); 13C NMR (150 MHz, [D6]DMSO, 258C): d=20.7,
21.0, 39.8, 41.1, 48.7, 52.7, 124.1, 125.9, 128.5, 129.0, 129.9, 132.0, 136.9,
ACHTUNGTRENNUNG(H,H)=8.4 Hz, 4H), 7.55 (AA’BB’ system, JACHTUNGTRENNUNG
3
AHCTUNGTRENNUNG
ACHTUNGTRENNUNG
138.5, 142.4, 143.9 ppm; IR (ATR): n˜ =3282, 2922, 1324, 1153 cmꢀ1
;
HRMS (ESI): m/z calcd for C40H44N4S4O8+Na+: 859.1934; found:
859.1899.
Cyclohexadiene 14: Column chromatography: from hexanes/dichlorome-
thane (6:4) to hexanes/dichloromethane/ethyl acetate (6:4:1) to afford 14
(0.03 g, 65%) as a colourless solid. M.p. 97–1018C; 1H NMR (600 MHz,
[D6]DMSO, 258C): d=1.30 (s, 9H), 2.27 (s, 3H), 2.33 (s, 3H), 2.35–2.40
(m, 1H), 2.35–2.40 (m, 1H), 2.42 (s, 3H), 2.45 (s, 3H), 2.50–2.51 (m, 1H),
Acknowledgements
Financial support from the Spanish MICINN (CTQ2008-05409-C02-02,
CTQ2008-03077, CTQ2011-23156, CTQ2011-23121 and CTQ2009-08328)
and the Catalan DIUE of the Generalitat de Catalunya (2009SGR637) is
acknowledged. A.D. thanks the Spanish MEC for a doctoral fellowship.
Support for the research of M.S. was received through the ICREA Aca-
demia 2009 prize for excellence in research funded by the DIUE. We
also acknowledge the Centre de Serveis Cientꢂfics i Acadꢈmics de Cata-
lunya (CESCA) for partial funding of computer time.
3.28–3.30 (m, 1H), 3.45 (dd, 3J
3.61 (app d, 2J
(H,H)=16.8 Hz, 1H), 3.79 (m, 1H), 3.81–3.83 (m, 1H),
3.85–3.87 (m, 3H), 3.90 (app d, 2J
(H,H)=16.8 Hz, 2H), 4.09–4.11 (m,
1H), 7.28 (BB’ system, 3J(H,H)=7.8 Hz, 2H), 7.30 (BB’ system, 3J-
(H,H)=7.8 Hz, 2H), 7.47 (BB’ system, 3J
(H,H)=7.8 Hz, 2H), 7.50 (BB’
system, 3J(H,H)=7.8 Hz, 2H), 7.54 (AA’ system, 3J
(H,H)=8.4 Hz, 2H),
7.57 (t, 3J(H,H)=6 Hz, 1H), 7.65 (AA’ system, 3J
(H,H)=8.4 Hz, 2H),
7.66 (AA’ system, 3J(H,H)=7.8 Hz, 2H), 7.81 ppm (AA’ system, 3J-
(H,H)=7.8 Hz, 2H); 13C NMR (150 MHz, [D6]DMSO, 258C): d=20.9,
21.0, 30.0, 41.0, 42.0, 43.2, 48.2, 49.8, 50.2, 54.9, 84.5, 122.2, 124.3, 126.6,
(H,H)=5.4 Hz, 2J
ACHUTGTNRENNGU ACHTUNGTERN(NUGN H,H)=14.4 Hz, 1H),
ACHTUNGTRENNUNG
AHCTUNGTRENNUNG
AHCTUNGTRENNUNG
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
A
ACHTUNGTRENNUNG
ACHTUNGTRENNUNG
A
[1] For reviews, see: a) K. P. C. Vollhardt, Angew. Chem. 1984, 96, 525;
Angew. Chem. Int. Ed. Engl. 1984, 23, 539; b) M. Lautens, W. Klute,
Chem. Eur. J. 2011, 17, 14493 – 14507
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14505

M. Solꢁ, A. Roglans et al.
W. Tam, Chem. Rev. 1996, 96, 49; c) P. R. Chopade, J. Louie, Adv.
Synth. Catal. 2006, 348, 2307; d) V. Gandon, C. Aubert, M. Malacria,
Chem. Commun. 2006, 2209; e) G. Domꢂnguez, J. Pꢉrez-Castells,
Chem. Soc. Rev. 2011, 40, 3430.
[6] a) P. A. Evans, K. W. Lai, J. R. Sawyer, J. Am. Chem. Soc. 2005, 127,
12466; b) T. Shibata, Y. Arai, Y. Tahara, Org. Lett. 2005, 7, 4955;
c) T. Shibata, M. Otomo, Y.-k. Tahara, K. Endo, Org. Biomol.
Chem. 2008, 6, 4296.
[7] a) T. Shibata, H. Kurokawa, K. Kanda, J. Org. Chem. 2007, 72, 6521;
b) K. Tanaka, G. Nishida, H. Sagae, M. Hirano, Synlett 2007, 1426;
c) Ref. [4g]; d) S. Brun, M. Parera, A. Pla-Quintana, A. Roglans, T.
Leꢊn, T. Achard, J. Solꢁ, X. Verdaguer, A. Riera, Tetrahedron 2010,
66, 9032.
[8] a) J. A. Varela, S. G. Rubꢂn, C. Gonzꢀlez-Rodrꢂguez, L. Castedo, C.
Saꢀ, J. Am. Chem. Soc. 2006, 128, 9262; b) J. A. Varela, S. Garcꢂa-
Rubꢂn, L. Castedo, C. Saꢀ, J. Org. Chem. 2008, 73, 1320; c) S.
Garcꢂa-Rubꢂn, J. A. Varela, L. Castedo, C. Saꢀ, Chem. Eur. J. 2008,
14, 9772; d) J. A. Varela, C. Saꢀ, J. Organomet. Chem. 2009, 694,
143.
[9] a) D. Leboeuf, L. Iannazzo, A. Geny, M. Malacria, K. P. C. Voll-
hardt, C. Aubert, V. Gandon, Chem. Eur. J. 2010, 16, 8904; b) V.
Gandon, N. Agelet, K. P. C. Vollhardt, M. Malacria, C. Aubert, J.
Am. Chem. Soc. 2006, 128, 8509.
[10] See footnote 20 in: Y. Shibata, Y. Otake, M. Hirano, K. Tanaka,
Org. Lett. 2009, 11, 689.
[11] J. J. Kaloko, Y-H. G. Teng, I. Ojima, Chem. Commun. 2009, 4569.
[12] a) A. Pla-Quintana, A. Roglans, A. Torrent, M. Moreno-MaÇas, J.
Benet-Buchholz, Organometallics 2004, 23, 2762; b) S. Cerezo, J.
Cortꢉs, D. Galvan, E. Lago, C. Marchi, E. Molins, M. Moreno-
MaÇas, R. Pleixats, J. Torrejꢊn, A. Vallribera, Eur. J. Org. Chem.
2001, 329.
[13] A. Dachs, S. Osuna, A. Roglans, M. Solꢁ, Organometallics 2010, 29,
562.
[14] For the cycloaddition of three acetylene molecules catalysed by
[2] For selected references involving two alkynes and one alkene, see:
a) D. W. Macomber, A. G. Verma, R. D. Rogers, Organometallics
1988, 7, 1241; b) S.-I. Ikeda, N. Mori, Y. Sato, J. Am. Chem. Soc.
1997, 119, 4779; c) N. Mori, S.-I. Ikeda, Y. Sato, J. Am. Chem. Soc.
1999, 121, 2722; d) V. Gandon, D. Leboeuf, S. Amslinger, K. P. C.
Vollhardt, M. Malacria, C. Aubert, Angew. Chem. 2005, 117, 7276;
Angew. Chem. Int. Ed. 2005, 44, 7114; e) G. Hilt, A. Paul, K. Harms,
J. Org. Chem. 2008, 73, 5187; f) Y. Obora, Y. Satoh, Y. Ishii, J. Org.
Chem. 2010, 75, 6046; for selected references involving diynes and
alkenes, see: g) Y. Wakatsuki, T. Kuramitsu, H. Yamazaki, Tetrahe-
dron Lett. 1974, 15, 4549; h) H. Suzuki, K. Itoh, Y. Ishii, K. Simon,
J. A. Ibers, J. Am. Chem. Soc. 1976, 98, 8494; i) Z. Zhou, L. P. Batta-
glia, G. P. Chiusoli, M. Costa, M. Nardelli, C. Pelizzi, G. Predieri, J.
Chem. Soc. Chem. Commun. 1990, 1632; j) Z. Zhou, M. Costa, G. P.
Chiusoli, J. Chem. Soc. Perkin Trans. 1 1992, 1399; k) E. S. Johnson,
G. J. Balaich, I. P. Rothwell, J. Am. Chem. Soc. 1997, 119, 7685; l) Y.
Yamamoto, H. Kitahara, R. Hattori, K. Itoh, Organometallics 1998,
17, 1910; m) T.-Y. Hsiao, K. C. Santhosh, K.-F. Liou, C.-H. Chen, J.
Am. Chem. Soc. 1998, 120, 12232; n) Y. Yamamoto, H. Kitahara, R.
Ogawa, K. Itoh, J. Org. Chem. 1998, 63, 9610; o) S.-I. Ikeda, H. Wa-
tanabe, Y. Sato, J. Org. Chem. 1998, 63, 7026; p) T. Sambaiah, L.-P.
Li, D.-J. Huang, C.-H. Lin, D. K. Rayabarapu, C.-H. Cheng, J. Org.
Chem. 1999, 64, 3663; q) Y. Yamamoto, H. Kitahara, R. Ogawa, H.
Kawaguchi, K. Tatsumi, K. Itoh, J. Am. Chem. Soc. 2000, 122, 4310;
r) T. Sambaiah, D.-J. Huang, C.-H. Cheng, J. Chem. Soc. Perkin
Trans. 1 2000, 195; s) M. J. Sung, J.-H. Pang, S.-B. Park, J. K. Cha,
Org. Lett. 2003, 5, 2137; t) Y. Yamamoto, S. Kuwabara, Y. Ando, H.
Nagata, H. Nishiyama, K. Itoh, J. Org. Chem. 2004, 69, 6697; u) M.-
S. Wu, D. K. Rayabarapu, C.-H. Cheng, Tetrahedron 2004, 60,
10005; v) S. Kezuka, S. Tanaka, T. Ohe, Y. Nakaya, R. Takeuchi, J.
Org. Chem. 2006, 71, 543; w) C. Aubert, V. Gandon, S. Han, B. M.
Johnson, M. Malacria, S. Schçmenauer, K. P. C. Vollhardt, G. D.
Whitener, Synthesis 2010, 2179.
[RhClACHTNUGTRENNUNG
H and Ph, whereas the Gibbs energy barrier for the rate-determin-
ing step is 19.8 and 24.7 kcalmolꢀ1 for R=H and Ph, respectively
(see Ref. [13]).
[15] A. Dachs, A. Torrent, A. Roglans, T. Parella, S. Osuna, M. Solꢁ,
Chem. Eur. J. 2009, 15, 5289.
[16] A. Dachs, A. Roglans, M. Solꢁ, Organometallics 2011, 30, 3151.
[17] Indeed, at the same level of theory, substitution of two phosphine li-
[3] For selected references involving enynes and alkynes, see: a) S.
Kezuka, T. Okado, E. Niou, R. Takeuchi, Org. Lett. 2005, 7, 1711;
for selected references involving enediynes, see: b) E. D. Sternberg,
K. P. C. Vollhardt, J. Org. Chem. 1982, 47, 3447; c) F. Slowinski, C.
Aubert, M. Malacria, Tetrahedron Lett. 1999, 40, 707; d) F. Slowin-
ski, C. Aubert, M. Malacria, Tetrahedron Lett. 1999, 40, 5849; e) F.
Slowinski, C. Aubert, M. Malacria, Adv. Synth. Catal. 2001, 343, 64;
f) F. Slowinski, C. Aubert, M. Malacria, Eur. J. Org. Chem. 2001,
3491; g) F. Slowinski, C. Aubert, M. Malacria, J. Org. Chem. 2003,
68, 378; h) P. Eckenberg, U. Groth, Synlett 2003, 2188; i) Ref. [2p];
j) M. Schelper, O. Buisine, S. Kozhuskov, C. Aubert, A. de Meijere,
M. Malacria, Eur. J. Org. Chem. 2005, 3000; k) A. Geny, S. Gaudrel,
F. Slowinski, M. Amatore, G. Chouraqui, M. Malacria, C. Aubert, V.
Gandon, Adv. Synth. Catal. 2009, 351, 271.
[4] a) R. Grigg, R. Scott, P. Stevenson, J. Chem. Soc. Perkin Trans. 1
1988, 1365; b) C. H. Oh, H. R. Sung, S. H. Jung, Y. M. Lim, Tetrahe-
dron Lett. 2001, 42, 5493; c) P. A. Evans, J. R. Sawyer, K. W. Lai,
J. C. Huffman, Chem. Commun. 2005, 3971; d) A. Torrent, I. Gon-
zꢀlez, A. Pla-Quintana, A. Roglans, M. Moreno-MaÇas, T. Parella, J.
Benet-Buchholz, J. Org. Chem. 2005, 70, 2033; e) B. Bennacer, M.
Fijiwara, S.-Y. Lee, I. Ojima, J. Am. Chem. Soc. 2005, 127, 17756;
f) I. Gonzꢀlez, S. Bouquillon, A. Roglans, J. Muzart, Tetrahedron
Lett. 2007, 48, 6425; g) S. Brun, L. Garcia, I. Gonzꢀlez, A. Torrent,
A. Dachs, A. Pla-Quintana, T. Parella, A. Roglans, Chem. Commun.
2008, 4339; h) L. X. Alvarez, B. Bessiꢈres, J. Einhorn, Synlett 2008,
1376; i) A. L. Jones, J. K. Snyder, J. Org. Chem. 2009, 74, 2907.
[5] a) K. Tsuchikama, Y. Kuwata, T. Shibata, J. Am. Chem. Soc. 2006,
128, 13686; b) T. Shibata, A. Kawachi, M. Ogawa, Y. Kuwata, K.
Tsuchikama, K. Endo, Tetrahedron 2007, 63, 12853; c) M. Kobaya-
shi, T. Suda, K. Noguchi, K. Tanaka, Angew. Chem. Int. Ed. 2011,
50, 1664.
gands by acetylene molecules in [RhClACTHUNGTRENNNUG
about 6 kcalmolꢀ1, whereas in [RhCl
11 kcalmolꢀ1 (see Ref. [13]).
ACHTUNGTRENNUNG
[18] a) W. Scherer, G. S. McGrady, Angew. Chem. 2004, 116, 1816;
Angew. Chem. Int. Ed. 2004, 43, 1782; b) T. S. Thakur, G. R. Desira-
ju, J. Mol. Struct. THEOCHEM 2007, 810, 143.
[19] L. Jiao, M. Lin, Z.-X. Yu, J. Am. Chem. Soc. 2011, 133, 447.
[20] a) A. D. Becke, J. Chem. Phys. 1993, 98, 5648; b) C. Lee, W. Yang,
R. G. Parr, Phys. Rev. B 1988, 37, 785; c) P. J. Stephens, F. J. Devlin,
C. F. Chabalowski, M. J. Frisch, J. Phys. Chem. 1994, 98, 11623.
[21] Gaussian 03, Revision C.01, M. J. Frisch, G. W. Trucks, H. B. Schle-
gel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, J. A. Montgom-
ery, 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, V. Bakken, C. Adamo, J. Jaramillo, R. Gomperts, R. E.
Stratmann, O. Yazyev, A. J. Austin, R. Cammi, C. Pomelli, J. W.
Ochterski, P. Y. Ayala, K. Morokuma, G. A. Voth, P. 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. Cio-
slowski, B. B. Stefanov, G. Liu, A. Liashenko, P. Piskorz, I. Komaro-
mi, 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, Inc., Wall-
ingford CT, 2004.
[22] P. Atkins, J. De Paula, Physical Chemistry, 8th ed., Oxford Universi-
ty Press, Oxford, 2006.
[23] C. Gonzalez, H. B. Schlegel, J. Chem. Phys. 1989, 90, 2154.
14506
www.chemeurj.org
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2011, 17, 14493 – 14507

Intramolecular Cycloadditions of Enediynes
FULL PAPER
[24] a) T. H. Dunning, Jr., J. Chem. Phys. 1989, 90, 1007; b) D. E. Woon,
T. H. Dunning, Jr., J. Chem. Phys. 1993, 98, 1358.
[25] K. A. Peterson, D. Figgen, M. Dolg, H. Stoll, J. Chem. Phys. 2007,
126, 124101.
[26] a) J. H. Hardesty, J. B. Koerner, T. A. Albright, G.-Y. Lee, J. Am.
Chem. Soc. 1999, 121, 6055; b) L. F. Veiros, G. Dazinger, K. Kirch-
ner, M. J. Calhorda, R. Schmid, Chem. Eur. J. 2004, 10, 5860.
[27] a) Q. Cui, D. G. Musaev, K. Morokuma, Organometallics 1997, 16,
1355; b) Q. Cui, D. G. Musaev, K. Morokuma, Organometallics
1998, 17, 742; c) Q. Cui, D. G. Musaev, K. Morokuma, Organometal-
lics 1998, 17, 1383; d) W. Zheng, A. Ariafard, Z. Lin, Organometal-
lics 2008, 27, 246; e) Y. Abe, K. Kuramoto, M. Ehara, H. Nakatsuji,
M. Suginome, M. Murakami, Y. Ito, Organometallics 2008, 27, 1736.
[28] L. Orian, J. N. P. van Strale, F. M. Bickelhaupt, Organometallics
2007, 26, 3816.
Received: July 19, 2011
Published online: November 23, 2011
Chem. Eur. J. 2011, 17, 14493 – 14507
ꢄ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
14507