Palladium(II)-catalyzed cycloisomerization of substituted 1,5-hexadienes: A combined experimental and computational study on an open and an interrupted hydropalladation/carbopalladation/β-hydride elimination (HCHe) catalytic cycle

Article abstract of DOI:10.1021/jo3004088

The Pd^II-catalyzed cycloisomerization of 3-alkoxycarbonyl-3- hydroxy-substituted 1,5-hexadienes has been studied experimentally and computationally. Experimentally, the reaction is characterized by a rapid room temperature formation of monomeric as well as dimeric cycloisomerization products using the commercially available precatalyst [(CH₃CN) ₄Pd](BF₄)₂. In situ NMR measurements indicate the initial kinetic advantage of the desired cycloisomerization pathway to methylene cyclopentanes; however, double bond isomerization, elimination, and dimer formation are competitive undesired pathways. Evaluation of the obtained product structures by NMR spectroscopy and X-ray crystallography indicates that the sole determinant for the monomer/dimer ratio is the regioselectivity of the initial hydropalladation in favor of the allylic (monomer formation) or the homoallylic double bond (dimer formation). In order to account for the experimental results, we propose the coexistence of two product-forming catalytic cycles, an open, monomer generating, as well as an interrupted and redirected, dimer generating, hydropalladation/carbopalladation/β-hydride elimination (HCHe) process. Results from computational studies of the proposed competing catalytic cycles are supportive to our mechanistic hypothesis and pinpoint the pivotal importance of Pd^II-hydroxo-chelate complexes for the reactivity-stability interplay of on- and off-pathway intermediates.

Full text of DOI:10.1021/jo3004088

Article
pubs.acs.org/joc
Palladium(II)-Catalyzed Cycloisomerization of Substituted
1,5-Hexadienes: A Combined Experimental and Computational
Study on an Open and an Interrupted Hydropalladation/
Carbopalladation/β-Hydride Elimination (HCHe) Catalytic Cycle
Bjorn Nelson,* Sonja Herres-Pawlis, Wolf Hiller, Hans Preut, Carsten Strohmann,
̈
and Martin Hiersemann*
Fakultat Chemie, Technische Universitat Dortmund, 44227 Dortmund, Germany
̈
̈
S
* Supporting Information
ABSTRACT: The PdII-catalyzed cycloisomerization of 3-alkoxycarbonyl-
3-hydroxy-substituted 1,5-hexadienes has been studied experimentally and
computationally. Experimentally, the reaction is characterized by a rapid
room temperature formation of monomeric as well as dimeric
cycloisomerization products using the commercially available precatalyst
[(CH3CN)4Pd](BF4)2. In situ NMR measurements indicate the initial
kinetic advantage of the desired cycloisomerization pathway to methylene
cyclopentanes; however, double bond isomerization, elimination, and
dimer formation are competitive undesired pathways. Evaluation of the
obtained product structures by NMR spectroscopy and X-ray
crystallography indicates that the sole determinant for the monomer/
dimer ratio is the regioselectivity of the initial hydropalladation in favor of
the allylic (monomer formation) or the homoallylic double bond (dimer
formation). In order to account for the experimental results, we propose the coexistence of two product-forming catalytic cycles,
an open, monomer generating, as well as an interrupted and redirected, dimer generating, hydropalladation/carbopalladation/
β-hydride elimination (HCHe) process. Results from computational studies of the proposed competing catalytic cycles are
supportive to our mechanistic hypothesis and pinpoint the pivotal importance of PdII-hydroxo-chelate complexes for the
reactivity−stability interplay of on- and off-pathway intermediates.
reported the cycloisomerization of substituted 1,5-hexadienes
by a cationic (phen)PdII-complex (Scheme 1, PdII⊕).16
In comparison to 1,6-heptadienes, 1,5-hexadienes are more
challenging substrates because they are generally less reactive
and frequently require elevated reaction temperatures to
enforce conversion. For instance, the cycloisomerization of
dimethyl allyl(propenyl)malonate by (t-BuCN)2PdCl2 required
forcing conditions (110 °C, 72 h), and five different cy-
cloisomerization products were detected (Scheme 1, PdII).17
Extensive experimental investigations by Lloyd-Jones revealed
that 1,5-hexadienes, which were formed as byproducts during
the cycloisomerization 1,6-heptadienes, act as powerful “Pd−H-
traps” and cycloisomerization inhibitors, even if present in
only catalytical amounts (<5%).18,19 A further restriction arises
from the experimental observation that under the conditions of
the cycloisomerization, the desired methylene cyclopentanes
are converted to methyl cyclopentenes by subsequent double-
bond isomerization.
INTRODUCTION
■
Substituted cyclopentanes represent a common structural
motive in secondary natural products and/or molecules with
interesting pharmacological properties. Therefore, a plethora of
synthetic strategies for the synthesis of cyclopentane building
blocks is available.1 The transition-metal-catalyzed cyclo-
isomerization of linear 1,n-dienes, -enynes, and -diynes is a
proven synthetic method for the construction of functionalized
carbocycles;2 the cycloisomerization is particularly attractive
because it creates, perfectly atom-economic, structural complex-
ity from easily available acyclic molecules of much lower
molecular complexity.3 2,2-Diallyl malonates have been used as
“benchmark” substrates for the metal-catalyzed cycloisomeriza-
tion of 1,6-dienes.4−9 In stark contrast, however, only a small
number of examples exist for the cycloisomerization of
1,5-hexadienes, although this reaction has its roots in the
middle of the last century.10−12 For instance, Livinghouse
has reported the cycloisomerization of a functionalized
1,5-hexadiene to the corresponding methylene cyclopentane
in the presence of a Ti-catalyst8b (Scheme 1, TiII). Examples for
the cycloisomerization of unsubstituted 1,5-dienes with late
transition-metals where initially revealed by Wilke,13 Keim,14
and Walther15 (nickel-catalysis). Widenhoefer subsequently
Despite the looming difficulties in terms of reactivity and
chemo- and regioselectivity, we embarked to establish conditions
Received: March 2, 2012
© XXXX American Chemical Society
A
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
1,5-hexadiene 1a20 to [Pd(MeCN)4](BF4)2 in CH2Cl2 at room
temperature, we obtained the desired cycloisomerization
product, methylene cyclopentane 2a, in encouraging yield
(Table 1, entry 1). Optimization of the reaction conditions in
terms of temperature (40 °C) and time (3.5 h) then delivered
2a with 68% yield, albeit with slightly eroded chemoselectivity
(2a/3a = 91/9) (Table 1, entry 2). The presence of molecular
sieves (ms) seems to efficiently suppress the cycloisomerization
reaction, even at elevated temperatures (Table 1, entry 3).22
Attempts to modulate the reactivity of the catalyst system by
the addition of monodentate phosphines (0.05 equiv of PCy3)
led to inferior yields and selectivities (Table 1, entry 4);
likewise, [Pd(dppp)(PhCN)2](BF4)2 was not competent to
catalyze the cycloisomerization (Table 1, entry 5); notably,
performing the reaction in toluene at reflux led to the formation
of significant amounts of the α-keto ester 4 by an oxy-Cope
rearrangement (Table 1, entry 6).23 Switching to (MeCN)2PdCl2,
no cycloisomerization product was obtained; again, at higher
reaction temperatures (70 °C) the formation of 4 could be
detected (Table 1, entries 7 and 8). These observations underline
the importance of the development of a low-temperature catalyst
system. In an attempt to accomplish the cycloisomerization of
1a using a purposeful in situ generated LnPd−H catalyst, an
equimolar mixture of (MeCN)2PdCl2 and Et3SiH according to
Lloyd-Jones17 was employed, unfortunately, without any notable
success (Table 1, entry 9). Recruitment of [Pd(η3-allyl)-
(MeCN)2](OTf),24 which was shown to be an active catalyst
for the cycloisomerization of 1,6-heptadienes,9i resulted in the
formation of only small amounts of the cycloisomerization
product 2a, even after prolonged heating (Table 1, entry 10).
Pd(η3-allyl)(MeCN)2](BF4),25 containing a weakly coordinating
counterion, triggered cycloisomerization but is less reactive and
selective than [Pd(MeCN)4](BF4)2 (Table 1, entries 2 and 11).
Conceptual Catalytic Cycle for the Cycloisomerization.26
Guided by the accepted mechanism for the PdII-catalyzed
cycloisomerization of 1,6-dienes,2,16,17 we initially proposed a
conceptual catalytic cycle, which is triggered by the in situ
Scheme 1. Cycloisomerization of 1,5-Dienes: Literature
Precedence
for the cycloisomerization of highly functionalized 1,5-hexadienes
1, featuring heterotopic double bonds, to methylene cyclo-
pentanes 2. Our initial study revealed the usefulness of the
commercially available catalyst [Pd(MeCN)4](BF4)2 (Scheme 1,
bottom),20,21 but the synthetic value of the procedure was
hampered by a close competition between various reaction
pathways.
In this article, we report the details of a combined
experimental and computational endeavor that was undertaken
to compile evidence and information on the mechanistic details
of the PdII-catalyzed cycloisomerization of 1,5-hexadienes 1.
RESULTS AND DISCUSSION
■
Experimental Studies. We embarked on this part of our
study with a catalyst screening; selected results for a single
substrate are summarized in Table 1. When subjecting the
a
Table 1. Screening Precatalysts and Conditions for the Cycloisomerization of 1a
b
yield (%)
c
entry
catalyst, additives (equiv)
solvent
temp (°C)
time (h)
(2a/3a)
d
1
[Pd(MeCN)4](BF4)2 (0.05)
CH2Cl2
CH2Cl2
(CH2Cl)2
(CH2Cl)2
CH2Cl2
toluene
rt
19
3.5
4
50 (95/5)
2
[Pd(MeCN)4](BF4)2 (0.05)
40
40
70
75
110
40
70
rt
68 (91/9)
no reaction
44 (86/14)
no reaction
3
[Pd(MeCN)4](BF4)2 (0.05), ms (4 Å)
[Pd(MeCN)4](BF4)2 (0.05), PCy3 (0.05)
[Pd(dppp)(PhCN)2](BF4)2 (0.05)
[Pd(dppp)(PhCN)2](BF4)2 (0.05)
(MeCN)2PdCl2 (0.05)
4
19
19
19
72
72
19
19
19
5
e f
,
6
(80)
7
(CH2Cl)2
(CH2Cl)2
CH2Cl2
CH2Cl2
CH2Cl2
no reaction
79 (0/30/70)
no reaction
efg
, ,
8
(MeCN)2PdCl2 (0.05)
9
(MeCN)2PdCl2 (0.05), Et3SiH (0.05)
[Pd(η3-allyl)(MeCN)2](OTf) (0.05)
[Pd(η3-allyl)(MeCN)2](BF4) (0.05)
f
10
11
60
60
10 (95/5)
46 (94/6)
a
Experiments conducted with c = 0.1 M. Reactions at elevated temperatures where run in a sealed tube. ms = molecular sieves; Cy = c-C6H11; dppp =
b
c
1,3-bis(diphenylphosphino)propane. Isolated yield after column chromatography. Determined by 1H NMR spectroscopy of the purified product.
d
e
Attempts to perform the reaction at 0 °C led to very low conversion rates, and a reduced solubility of the catalyst was noticed. 4, product of an
f
g
1
oxy-Cope-rearrangement, was detected. Determined by H NMR spectroscopy of the crude reaction mixture. Ratio: (2a/3a/4).
B
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
formation of a putative (CH3CN)3PdII−H species from
[Pd(MeCN)4](BF4)2 (Scheme 2). Regioselective coordination
most electrophilic precatalyst, the “bis-cationic” [Pd(MeCN)4]-
(BF4)2 complex requires the shortest induction period.28 We
continued the kinetic profiling by monitoring the entire product
genesis from ( )-syn-1b in the presence of [Pd(MeCN)4](BF4)2
(0.05 equiv) in CDCl3 at 27 °C (Figure 1B). Formation of the
cyclopentane 2b is dominating from the outset and levels at 60%
after 10 h. Upon its formation, 2b is steadily consumed because
of double bond isomerization to 3b, which in turn, is
subsequently converted to cyclopentadiene 5b by formal
elimination of water. Notably, build-up of the previously unnoticed
dimer 6b29,30 was detected, and, when considering that the
connection of the diene fragment to the cyclopentane ring occurs at
C2′, it is evident that the genesis of 6b requires a different and
competing cycloisomerization pathway, which is initiated by
hydropalladation of the homoallylic double bond. Following the
maximum build-up (19%, 8 h), 6b is consumed by a downstream
bond building to products that were untraceable in this experiment.
Kinetic profiling of the solvent-dependence of the product
genesis led to further insight into the unprecedented
dimerization. Monitoring product formation for 1b in the
presence of [Pd(MeCN)4](BF4)2 (0.05 equiv) in acetone-d6
(27 °C, 20 h) (Figure 1C, 1D) indicates faster consumption
than in CDCl3 and decreased initial selectivity in favor of 2b
(44% maximum build-up after 8 h). As it was the case in
CDCl3, double bond isomerization to 3b and subsequent
elimination to 5b are notable. Furthermore, the competing
dimerization is more pronounced, which leads to a significant
build-up of 6b (33% after 6 h). Monocyclic dimer 6b is than
consumed to afford the dimer 7b;31 this bond forming cascade
is operative even after full consumption of 1b, indicating the
formation of 7b from 1b via 6b.
Generalized Product Distribution in 1,5-Hexadiene
Cycloisomerizations. The interplay and dependence of the
various bond-forming events under the cycloisomerization
conditions is subtle, and we expected the substrate structure
as a key determinant. In order to validate that the established
bond forming cascades are general, different 1,5-hexadienes
were synthesized and subjected to the conditions of the
cycloisomerization (Table 2). In line with our previous
experiments, the monocycloisomerized dimer 6c (22%) resulted
from the cycloisomerization of ( )-1c20 (Table 2, entry 1), and
the dimers 6d and 7d were obtained from the cyclo-
isomerization of ( )-1d (Table 2, entry 2). Notably, the
cycloisomerization of ( )-1e delivered the elusive regioiso-
meric methylene cyclopentane ( )-2e′, although in very minute
amounts, along with the dimer 7e as a mixture of diastereomers
(Table 2, entry 3); from this mixture, the S2 symmetric
diastereomer 7e could be crystallized, and crystal structure
analysis corroborated our structural assignment.32 In accord-
ance with a Thorpe−Ingold effect,33 the cycloisomerization of
( )-1f proceeded extremely fast (complete consumption of 1f
in less than 30 min) and afforded the expected mixture
monomers and dimers (Table 2, entry 4); the S2 symmetric
dimer 7f was subjected to X-ray crystallography.34
Scheme 2. Conceptual Catalytic Cycle for the
Cycloisomerization by a Hydropalladation/
Carbopalladation/β-Hydride Elimination (HCHe)
a
Sequence
a
Formation of the experimentally observed monomers 2, 3 and the
elimination product 5. E = CO2i-Pr, L = CH3CN.
of the cationic (CH3CN)3PdII−H catalyst to the allylic double
bond of 1 then delivers the (CH3CN)2PdII-π-complex I and
subsequent hydropalladation provides the (CH3CN)2PdII-σ-
alkyl-complex II. Ensuing 5-exo-trig carbopalladation of the
homoallylic double affords the (CH3CN)2PdII-σ-cycloalkyl-
intermediate III, which undergoes β-hydride elimination to
deliver the experimentally observed methylene cyclopentane 2;
double bond isomerization and elimination (vide infra) are
initiated by π-complex formation (IV) and subsequent
hydropalladation of the exocyclic double bond of 2 to deliver
the (CH3CN)2PdII-σ-complex V; ensuing β-hydride elimination
then provides methyl cyclopentene 3. In terms of ΔΔG°,
DFT calculations27 predict 3 to be significantly more stable
(−3.8 kcal/mol) than 2; consequently, we expect a build-up of
3 (and 5) on the expense of 2 as the reaction proceeds under
the conditions of the cycloisomerization.
Kinetic Experiments. In order to collect evidence in
support of the proposed catalytic cycle (Scheme 2), we
1
monitored the course of the cycloisomerization by in situ H
NMR spectroscopy (Figure 1). Using the 1,5-hexadiene ( )-syn-
1b, we first studied the rate and selectivity of different PdII-based
precatalysts in the formation of methylene cyclopentane 2b in
CDCl3 at 40 °C (Figure 1A). In accordance with our qualitative
experimental observations, [Pd(MeCN)4](BF4)2 stood out as
most reactive precatalyst. In the event, a maximum build-up of
2b (51%) was achieved after 45 min. The subsequent
consumption of 2b can be related to the formation of 3b by
double bond isomerization or 5b by elimination. The addition of
PCy3 (0.05 equiv) clearly decelerates the rate of formation of 2b
and the build-up of 2b levels at 46% after 18.5 h. The
performance of precatalysts of the type Pd(η3-allyl)(MeCN)2]-
(X) depends on the nature of the counterion X. For X = BF4,
55% of 2b were detected after 40 h, whereas for
X = OTf, only 32% 2b were formed within the duration of
the experiment (88 h). The S-shape of the kinetic profile
depicted in Figure 1A indicates a precatalyst-dependent
induction period for the formation of the actual cyclo-
isomerization catalyst, supposedly a LnPd−H species, and the
Conceptual Catalytic Cycle for Dimer Formation.
Considering the experimental evidence discussed above, a
feasible qualitative catalytic cycle for the formation of the dimers
6, 7 and the monomer 2′ from hexadiene 1 can be proclaimed
(Scheme 3). We propose a bond-forming cascade initiated by
an interrupted cycloisomerization of 1, which proceeds via the
hydropalladation product VI to the carbopalladation product
VII, which resists β-hydride elimination. As an alternative for
β-hydride elimination, formation of the π-complex VIII can be
C
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 1. (A) Screening (1H NMR, CDCl3, 40 °C) of precatalysts (0.05 equiv) for the cycloisomerization of 1b to 2b. (B) Kinetic profile (1H NMR,
CDCl3, 27 °C) for the cycloisomerization of 1b with [Pd(MeCN)4](BF4)2 (0.05 equiv). (C) 1H NMR kinetic profile (1H NMR, acetone-d6, 27 °C)
for the cycloisomerization of 1b with [Pd(MeCN)4](BF4)2 (0.05 equiv). (D) Section of the 1H NMR array. All experiments were performed using a
Shigemi NMR tube and with c = 0.1 M.
assumed, and subsequent intermolecular carbopalladation to
afford IX redirects the initial cycloisomerization toward
dimerization. The dimerization process is continued by
β-hydride elimination to afford the π-complex X and the
experimentally observed monocyclic dimer 6. Subsequent
hydropalladation of 6 at the terminal double bond delivers the
PdII-σ-complex XI, which undergoes an intramolecular
carbopalladation and subsequent β-hydride elimination to
conclude the catalytic cycle and to provide the experimentally
observed bicyclic dimer 7.
Interrupted and Redirected Cycloisomerization. Ac-
cording to our mechanistic proposal, the β-hydride elimination
from the Pd(II)-σ-cycloalkyl-intermediate III is essential to
conclude the simple catalytic cycle of the HCHe cascade
(Scheme 2). Consequently, it should be possible to telescope
the cycloisomerization pathway, which was initiated by hydro-
palladation of the allylic double bond, toward dimerization by
preventing β-hydride elimination; in this regard, replacement
of the β-hydride by a non-hydrogen atom substituent should
enforce dimerization by Heck reaction (Scheme 4). In order
to experimentally validate the proposal of an interrupted and
redirected cycloisomerization (Scheme 3), the tailor-made
1,5-hexadiene ( )-1g was subjected to [Pd(MeCN)4](BF4)2
(0.05 equiv) in CH2Cl2 at room temperature for 72 h and
delivered the expected monocycloisomerized dimer ( )-8 as a
mixture of diastereomers.
COMPUTATIONAL CHEMISTRY
■
DFT Calculations. Computational studies were initiated with the
intent to support and to refine the conceptual catalytic cycle for the
cycloisomerization. In particular, we sought to understand how the
regioselectivity of the initial hydropalladation (allylic versus homo-
allylic double bond) predetermines the final product formation
(monomers versus dimers). According to the conceptual catalytic
cycles (Schemes 2 and 4), the overall cycloisomerization process
consists of a hydropalladation/carbopalladation/β-hydride elimination
(HCHe) sequence; accordingly, the results of the DFT calculations
will be presented cumulatively.
Computational Details. All calculations where performed with
the Gaussian0335 suite of programs at DFT level with the B3LYP36
functional. Geometry optimizations were performed without re-
strictions and default convergence criteria with the LACVP(d)37
basis set in the gas phase at 298.15 K. The basis set consists of
the standard 6-31G(d) basis set for the atoms (H, C, N, O) and the
LANL2DZ basis set, which contains an effective core potential (ECP)
D
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
a
Table 2. Product Distribution in the Cycloisomerization of Substituted 1,5-Hexadienes 1c−f
a
b
1
Conditions: [Pd(MeCN)4](BF4)2 (0.05 equiv), CH2Cl2 (c = 0.1 M), rt, 3−23 h. E = CO2i-Pr. Determined by H NMR spectroscopy of the
c
d
purified product mixture. Addition of PCy3 (0.05 equiv) in order to induce a more manageable rate of conversion. Ratio: (2/2′/3/6/7).
Scheme 3. Conceptual Catalytic Cycle for the Interrupted
and Redirected Cycloisomerization
Scheme 4. Interrupted and Redirected Cycloisomerization:
Enforced Dimerization of 1,5-Hexadiene ( )-1g
a
a
Formation of the experimentally observed dimers 6, 7 and the
monomer 2′. E = CO2i-Pr, L = CH3CN.
ester group; (ii) the substituent at C4 was reduced to a methyl group;
and (iii) the catalytical active “Pd−H“-complex was expressed by
[PdH(MeCN)3]+ with square planar geometry. Energies reported in
the text section correspond to the relative electronic energies after
zero-point energy correction (ΔE+ZPC). In Figures 2−7, Gibbs free
energies (ΔG) are included in parentheses.
Regioselectivity of the Hydropalladation. The cycloisomerization-
triggering hydropalladation of the allylic or the homoallylic double
bond of the 1,5-hexadienes 1 constitutes the branching point for
subsequent product formation by the HCHe sequence (Scheme 5).
Therefore, we embarked on our computational endeavor by modeling
the competing hydropalladation pathways.
After a careful examination of various (CH3CN)2Pd·1,5-hexadiene
π-complexes, we identified the π-complex M1 as most stable bis(acetonitrile)
complex and normalized its energy to zero. Starting from M1, our gas
phase calculations predict that the hydropalladation of the allylic
double bond TS(M1−M2) proceeds almost barrierless and provides
the more stable (−4.2 kcal/mol) β-agostic complex M2 (Figure 2). Notably,
with a split valence (double-ζ) basis set for Pd. For more accurate
energy values, all optimized structures were reoptimized with the
def2-SVP38 (double-ζ) basis set for Pd and the 6-31++G(d,p) basis set
for the lighter atoms, as implemented in Gaussian03. Harmonic
frequencies were calculated at the same level of theory to identify the
nature of stationary points as minima (zero imaginary frequencies in
the Hessian matrix) or transition states (one imaginary frequency;
animation of the vibrational mode verified the reaction coordinate)
and to obtain zero-point energy corrections (ZPC). In addition,
relevant transitions states were verified by IRC-calculations. On the
optimized geometries, single-point calculations were performed with
the def2-TZVP38 39 (triple-ζ) basis set, defined by Weigend and
,
Ahlrichs. The obtained results showed no significant variations of the
previously observed trends. To reduce computational cost, we
employed a model system for our calculations with the following
simplifications: (i) the isopropyl ester group was replaced by a methyl
E
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
complexes led to the localization of the six-membered PdII-hydroxo-
chelate complex M7 and the seven-membered PdII-carbonyl complex
M8. No attempt was made to model the multiple conformational changes
required for the exoenergetic formation of M7 (−11.7 kcal/mol) from
M6. Nevertheless, we assume the kinetic accessibility of M7 and consider
M7 as inevitable intermediate on the (interrupted) HCHe pathway.
Overall, our analysis of the regioselectivity of the hydropalladation
suggests a rapid and irreversible process, which affords the PdII-hydroxo-
chelate complexes M3 and M7. In accordance with the experimentally
observed product ratios, the calculations do not rule out a competition
between the allylic and homoallylic hydropalladation pathways.
Carbopalladation and β-Hydride Elimination. Beyond the
hydropalladation branching point to the constitutional isomers M3
and M7, modeling the HCHe pathway requires separate treatment of
the fate of the two pivotal intermediates M3 and M7. Furthermore, for
both intermediates, the 5-exo-trig carbopalladation is a diastereoface-
differentiating elementary reaction. The results of the computational
analysis of the Re- and Si-face 5-exo-trig carbopalladation of the
homoallylic double bond in M3 will be discussed first.
Figure 2. B3LYP/(Def2-TZVP + 6-31++G(d,p)) calculated ΔE (ΔG)
in kcal/mol for the allylic hydropalladation; all energies relative to M1.
E = CO2Me, L = CH3CN.
The 5-exo-trig carbopalladation of the homoallylic double in M3 is,
as schematized in Scheme 6, a branching point to diastereomeric
Scheme 6. Principle Branching Point within the
a
Hydropalladation/Carbopalladation/β-Hydride Elimination
Scheme 5. Allylic vs. Homoallylic Hydropalladation
a
(HCHe) Process
a
E = CO2Me, L = CH3CN.
our initial hydropalladation scenario (Scheme 5) is expanded by the
localization of the five-membered PdII-hydroxo-chelate complex M3 as well
as the six-membered PdII-carbonyl-chelate complex M4.17,40 Considering the
low barrier (3.1 kcal/mol) for the formation of M3 from M2 and its stability
relative to M2 (−12.4 kcal/mol) and M4 (−3.8 kcal/mol), we consider M3
as pivotal intermediate on the reaction coordinate.
a
Re- and Si-Face carbopalladation of the homoallylic double bond.
E = CO2Me, L = CH3CN.
The competing hydropalladation of the homoallylic double bond
was modeled next (Figure 3). In the event, formation of the less stable
cyclopentylmethyl-PdII-σ-complexes, which by way of β-hydride
elimination, are converted to diastereomeric hydrido-PdII-π-complexes
containing the identical methylene cyclopentane ligand. Accordingly,
whether or not the 5-exo-trig carbopalladation is effectively diastereo-
face-differentiating, the existence of a principle branching point is
inconsequential, at least with respect to the expectable product
spectrum. However, and possibly unexpectedly, our computational
analysis outlined below suggests a more sophisticated course in which
kinetics prevent the formation of a thermodynamically dominant,
possibly HCHe-interrupting PdII-hydroxo-chelate complex.
The computationally predicted pathway required for and resulting
from the 5-exo-trig Si-face carbopalladation of M3 is depicted in Figure 4.
The carbopalladation event requires the formation of the π-complex
M9 from the PdII-hydroxo-chelate complex M3. The calculations predict an
endoenergetic intramolecular ligand exchange process (+4.7 kcal/mol)
with a significant barrier (12.1 kcal/mol). Notably, M9 appears to be a
1,3-allylic strain-minimized conformer. Subsequent 5-exo-trig carbo-
palladation proceeds from M9 via the cis-bicyclo[3.2.0]heptane-
like transition-state structure TS(M9−M10) with significant barrier
(+11.5 kcal/mol) to deliver exoenergetically (−1.2 kcal) the γ-agostic
complex M10. The predicted γ-agostic complex M10 is separated from
the significantly more stable (−8.5 kcal/mol) β-agostic complex M11
by a rather minute barrier (3.0 kcal/mol). Our search for chelate
complexes of the carbopalladation product led to the localization
of the PdII-carbonyl-σ-complex M12 (−21.2 kcal/mol relative to
M1), which is nearly isoenergetic to the β-agostic complex M11
(−21.6 kcal/mol relative to M1). The concluding β-hydride
Figure 3. B3LYP/(Def2-TZVP + 6-31++G(d,p)) calculated ΔE (ΔG)
in kcal/mol for the homoallylic hydropalladation; all energies relative
to M1. E = CO2Me, L = CH3CN.
(+2.6 kcal/mol) π-complex M5 from M1 is required to open a
barrierless exoenergetic (−3.2 kcal/mol) hydropalladation pathway
to the β-agostic σ-complex M6. Our search for stable chelate
F
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 4. B3LYP/(Def2-TZVP + 6-31++G(d,p)) calculated ΔE (ΔG) in kcal/mol for the Si-face 5-exo-trig carbopalladation of the homoallylic
double bond, and subsequent β-hydride elimination; all energies relative to M1. E = CO2Me, L = CH3CN.
Figure 5. B3LYP/(Def2-TZVP + 6-31++G(d,p)) calculated ΔE (ΔG) in kcal/mol for the Re-face 5-exo-trig carbopalladation of the homoallylic
double bond, and subsequent β-hydride elimination; all energies relative to M1. E = CO2Me, L = CH3CN.
elimination process proceeds endoenergetically and with a low barrier
from M11 via TS(M11−M13) to the hydrido-PdII-π-complex M13.41
Overall, our modeled pathway for a sequential 5-exo-trig carbopalla-
dation/β-hydride elimination from M3 to M13 is well passable with an
energetic span of 16.2 kcal/mol between the rate determining
intermediate M3 and the rate-determining transition-state TS(M9−
M10). The carbopalladation event is slow and irreversible, whereas the
concluding β-hydride elimination is fast and reversible. The nonproductive
PdII-carbonyl-σ-chelate complex M12 is energetically accessible but does
not represent an interrupting resting state for the HCHe process.
The course of competing 5-exo-trig Re-face carbopalladation of the
homoallylic double bond is depicted in Figure 5. The carbopalladation
elementary reaction requires the endoenergetic formation (+8.1 kcal/mol)
of the π-complex M15 from the PdII-hydroxo-chelate complex M3.
Notably, M15 could be destabilized by complexation-enforced 1,3-
allylic strain. In quite close analogy to the processing of M9 (Figure 4),
the carbopalladation of M15 proceeds with a manageable barrier to
provide, exoenergetically (−3.5 kcal/mol), the γ-agostic complex M16,
which is separated from the more stable β-agostic complex M17 by
insignificant barrier (2.8 kcal/mol). The marked difference between
G
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
the Re- and the Si-face carbopalladation process surfaced once we
located the hydroxo-chelate complex M18. Whereas the PdII-carbonyl-
chelate complex M12 is isoenergetic to the β-hydride complex M11
(Figure 4), the corresponding six-membered PdII-hydroxo-chelate
complex M18 represents a thermodynamic resting state, which renders
the concluding β-hydride elimination of the HCHe process
prohibitively endoenergetic (+14.2 kcal/mol).
Altogether, our calculations predict that the Re-face carbopallada-
tion of the homoallylic double would inevitably lead to the formation
of the PdII-hydroxo-chelate complex M18, a thermodynamic resting
state off the HCHe pathway, which would interrupt and, possibly,
redirect the HCHe pathway toward dimerization. Experimentally,
however, we have not yet been able to isolate noticeable amounts of
dimers resulting from a Heck-type reaction of M18. Consistently, our
calculations predict a kinetic preference for the Si-face over the Re-face
carbopalladation process (TS(M9−M10) versus TS(M15−M16),
ΔΔE‡ = 3 kcal/mol) and, therefore, in favor of the uninterrupted
HCHe pathway.
So far, we revealed that in accordance with the experimentally
observed formation of methylene cyclopentanes 2a−f by the cy-
cloisomerization of 1,5-hexadienes 1a−f, modeling the HCHe process
initiated at the allylic double bond suggests a passable pathway from
the π-complex M1 via M3 and M11 to M13. Having already calculated
the hydropalladation pathway from M1 to M7 (Figure 5), we focused
on the branching pathways arising from the Re- or Si-face 5-exo-trig
carbopalladation of the allylic double bond of M7 (Scheme 7).
Looking back at our conceptual catalytic cycle for dimer forma-
tion (Scheme 4), the then proposed resistance of the concep-
tual intermediate VII toward β-hydride elimination can now be
understood as a consequence of the formation of off-pathway
PdII-O-donor-chelate σ-complexes, M24 or M25 (Figure 6), whose
thermodynamic stability interrupts the HCHe catalytic cycle and
redirects the bond forming cascade toward dimer formation. This
mechanistic origin of the experimentally observed dimer formation
is even more convincingly illustrated by the modeled pathway of
the Si-face 5-exo-trig carbopalladation of the allylic double bond
(Figure 7).
The modeled pathway initiated by the Si-face 5-exo-trig carbopallada-
tion of the allylic double bond progresses from the PdII-hydroxo-chelate
complex M7 via the π-complex M27 and the transition state TS(M27−
M28) to provide, highly exoenergetic (−20.2 kcal/mol), the five-
membered PdII-hydroxo-chelate complex M28; upstream agostic PdII
σ-complexes could not be located. Comparing the calculated activation
barriers for the competing carbopalladation pathways via TS(M21−
M22) (+15.9 kcal/mol, Figure 6) or TS(M27−M28) (+12.0 kcal/mol,
Figure 7) indicates a marked kinetic preference (ΔΔE‡ = 3.9 kcal/mol)
for the Si-face 5-exo-trig carbopalladation pathway depicted in Figure 7.
Consequently, we consider the Si-face 5-exo-trig carbopalladation
pathway as kinetically dominant. Hitherto, the transition state
of the carbopalladation elementary reaction represented the rate-
determining transition state (Figures 4−6). The predicted stability
of the PdII-hydroxo-chelate complex M28 now deeply changes the
appearance of the energy landscape (Figure 7): M28 at −31.9 kcal/mol
and the transition state for β-hydride elimination TS(M30−M31) at
−11.2 kcal/mol are now the energetic span-defining states, and the
resulting effective barrier (+20.7 kcal/mol) is much higher than
before. More significantly, the PdII-hydroxo-chelate complex M28
represents a prototypic off-pathway resting state that resists β-
hydride elimination and serves as redirecting point to the experi-
mentally noticed dimer formation. Notably, the predicted all-cis
relative configuration of M28 matches the relative configuration of
the S2 symmetric dimers 7e and 7f that were characterized by X-ray
crystallography.
Scheme 7. Principle Branching Point within the
Hydropalladation/Carbopalladation/β-Hydride Elimination
a
(HCHe) Process
CONCLUSION
■
Using the commercially available precatalyst [Pd(MeCN)4]-
(BF4)2, highly functionalized 1,5-hexadienes have been
converted to valuable methylene cyclopentane building blocks.
This low-temperature cycloisomerization is fast and effective;
double bond isomerization and elimination can be minimized
by optimizing the reaction conditions, but its efficiency is
clearly affected by the propensity for dimer formation. Catalyst
optimization has so far met with limited success; for instance,
the presence of phosphine ligands is detrimental to the activity
of the catalyst. In situ NMR studies delivered valuable insights
into the product formation process, and X-ray crystallography
helped to assign the constitution and configuration of dimeric
products.
To summarize the computational part of this study, the
predicted essentials of the competing open and interrupted
hydropalladation/carbopalladation/β-hydride elimination
(HCHe) processes are depicted in Figure 8. The site-selectivity
of the initiating hydropalladation determines the chemo-
selectivity (monomer versus dimer formation). Hydropallada-
tion at the allylic double bond initiates an intact HCHe
cycle, which proceeds from M1 by a fast and irreversible
hydropalladation to deliver the PdII-hydroxo-chelate complex
M3, which is converted by the rate-limiting and irreversible
carbopalladation to the β-agostic methylcyclopentane PdII
σ-complex M11. Fast and reversible β-hydride elimination then
affords the π-complex M13, and the slightly exoenergetic ligand
exchange to M1 completes the HCHe cycle. The hydropalladation
a
Re- and Si-Face carbopalladation of the allylic double bond. E =
CO2Me, L = CH3CN.
Intrigued by the unexpected experimental observation that hydro-
palladation of the homoallylic double bond triggers formation of
dimers rather than monomeric methylene cyclopentanes (Scheme 4),
hence disapproving the expected (Scheme 7), we sought mechanistic
insights by computational chemistry.
The computed pathway for the Re-face 5-exo-trig carbopalladation
of the allylic double bond will be discussed first (Figure 6). Starting
from the PdII-hydroxo-chelate complex M7 and with the inter-
mediacy of the conformer M20, we propose a rapid formation of the
PdII-olefin-chelate π-complex M21. The elementary carbopalladation
process proceeds slightly endonergetically (+1.7 kcal/mol from M7)
and with a conquerable barrier (15.0 kcal/mol) to deliver the γ-agostic
complex M22. In line with our previous computations, the γ-agostic
complex is separated from the significantly more stable β-agostic
complex M23 (−8.4 kcal/mol) only by a low-lying transition state
(+3.1 kcal/mol). Mindful of the importance of PdII-O-donor-chelate
complexes in shaping the energy landscape, we scanned for and
localized the off-pathway chelate complexes M24 (−5.3 kcal/mol
relative to M23) and M25 (−6.6 kcal/mol relative to M23), both
markedly more stable than the on-pathway complex M23.
Consequently, the stability and the proposed kinetic accessibility of
M25 would interrupt the HCHe process by rendering the terminating
β-hydride elimination to M26 unfavorable (+10.7 kcal/mol).
H
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 6. B3LYP/(Def2-TZVP + 6-31++G(d,p)) calculated ΔE (ΔG) in kcal/mol for the Re-face 5-exo-trig carbopalladation of the allylic double
bond, and subsequent β-hydride elimination; all energies relative to M1. E = CO2Me, L = CH3CN.
Figure 7. B3LYP/(Def2-TZVP + 6-31++G(d,p)) calculated ΔE (ΔG) in kcal/mol for the Si-face 5-exo-trig carbopalladation of the allylic double
bond, and subsequent β-hydride elimination; all energies relative to M1. E = CO2Me, L = CH3CN.
of the homoallylic double bond is kinetically competitive and ini-
tiates an interrupted and redirected HCHe cycle, which proceeds
from M5 by a fast and irreversible hydropalladation to the
PdII-hydroxo-chelate complex M7, which is converted by
carbopalladation to afford the off-pathway PdII-hydroxo-chelate
complex M28, whose intermediacy renders the β-hydride elimi-
nation to M31 prohibitively endoenergetic. We propose that
the PdII-hydroxo-chelate complex M28 represents the interrupting
off-pathway intermediate, which redirects the HCHe pathway
toward dimer formation.
We believe that as soon as the complexes M1 or M5 are
formed, the hydropalladation occurs rapidly and irreversibly;
the site-selectivity and, therefore, chemoselectivity originates
from the capability of the catalyst to differentiate between
the heterotopic double bonds. The current catalyst system
is moderately successful in this regard, but our computations
I
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
Figure 8. Computationally predicted essentials of the competition between the open and the interrupted hydropalladation/carbopalladation/
β-hydride elimination (HCHe) processes. E = CO2Me, L = CH3CN.
provide no insights into factors that determine the site-
differentiation. Future experimental work will be directed
toward the development of more selective, site-differentiating
hydropalladation catalysts that are competent to carry an open
HCHe cycle without surrendering the characteristics of the
substrate structure. Future efforts in computational chemistry
will be directed toward an understanding of the factors that
should promote the site-differentiating hydropalladation as well
as the mechanistic details of the dimer formation.
1,5-Hexadiene ( )-1e. According to the general procedure for
the [2,3]-Wittig rearrangement, the reaction of allyl vinyl ether 1045
(0.99 g, 5 mmol) with (i-Pr)2NH (0.84 mL, 6 mmol, 1.2 equiv) and
n-BuLi (2.4 mL, 5.5 mmol, 1.1 equiv, 2.3 M in hexanes) in THF
(10 mL + 15 mL) at −85 → −70 °C (10 min) and 0 °C (10 min)
provided 1,5-hexadiene ( )-1e (0.47 g, 2.4 mmol, 48%) after flash
chromatography (cyclohexane/ethyl acetate 200/1 → 50/1) as a
mixture of diastereomers (syn/anti = 77/23): Rf 0.66 (cyclohexane/
ethyl acetate 5/1); 1H NMR (CDCl3, 400 MHz, mixture of
diastereomers syn/anti = 77/23, δ) 0.99 (d, J = 6.9 Hz, 3H), 1.22−
1.31 (m, 6H), 2.54−2.69 (m, 1H), 3.27 (s, 0.8Hmajor), 3.31 (s,
0.2Hminor), 4.95−5.05 (m, 2H), 5.06−5.12 (m, 1H), 5.18 (dd, J = 10.5,
1.4 Hz, 0.2Hminor), 5.22 (dd, J = 10.5, 1.5 Hz, 0.8Hmajor), 5.45 (dd, J =
17.1, 1.4 Hz, 0.2Hminor), 5.52 (dd, J = 17.0, 1.5 Hz, 0.8Hmajor), 5.69−
5.81 (m, 1H), 5.89 (dd, J = 17.0, 10.5 Hz, 0.8Hmajor), eclipsed by 5.92
(dd, J = 17.2, 10.2 Hz, 0.2Hminor); 13C NMR (CDCl3, 101 MHz,
mixture of diastereomers syn/anti = 77/23, δ) 13.7/14.4 (CH3), 21.8/
21.9 (2 × CH3), 44.8/45.1 (CH), 70.2/70.3 (CH), 79.3/79.6 (C),
115.6/116.0 (CH2), 116.1/116.4 (CH2), 137.8/138.2 (CH), 138.4/
138.7 (CH), 174.4/174.5 (C); IR (cm−1) 3510(br s) (ν OH), 3080(m),
2980(s), 2940(m), 2875(m), 1725(s) (ν CO), 1640(s) (ν CC),
1470(s), 1455(s), 1340(m), 1375(s), 1260(s) (ν C−O−C ester),
1165(s), 1105(s), 995(s); HRMS (ESI) Calcd. for C11H18O3Na ([M +
Na]+) 221.11482, found 221.11485; C11H18O3, M = 198.26 g/mol.
EXPERIMENTAL SECTION
■
General Procedure for the Dienolate [2,3]-Wittig Rearrange-
ment.42 A cooled (−85 °C) solution of LDA [prepared in situ from
(i-Pr)2NH (1.2−1.3 equiv) and n-BuLi (1.1−1.4 equiv of n-BuLi,
titrated employing diphenylacetic acid as indicator43)] in dry THF
(2 mL/mmol) was treated dropwise with a cooled (−85 °C) solution
of the allyl vinyl ether (1 equiv, mixture of double bond isomers) in
dry THF (3 mL/mmol). The resulting light yellow solution was stirred
for a period of 10−30 min (−85 → approximately −70 °C). The dry
ice bath was then replaced by a cooling bath, and the reaction mixture
was stirred at 0 °C until TLC indicated complete conversion of the
starting material (10 min−1 h). In most cases, successful reaction was
accompanied by a color change from light yellow to orange/red. The
solution was subsequently diluted by the addition of aqueous saturated
NH4Cl solution at 0 °C and warmed to room temperature. The
aqueous layer was extracted with CH2Cl2 (3×). The combined organic
phases were dried (MgSO4) and concentrated under reduced pressure.
Purification by flash chromatography (cyclohexane/ethyl acetate)
afforded the corresponding 1,5-hexadiene ( )-1 as a mixture of
diastereomers.
1,5-Hexadiene ( )-1d. According to the general procedure for the
[2,3]-Wittig rearrangement, the reaction of allyl vinyl ether 944 (1.8 g,
10 mmol) with (i-Pr)2NH (1.8 mL, 13 mmol, 1.3 equiv) and n-BuLi
(5.1 mL, 12 mmol, 1.2 equiv, 2.35 M in hexanes) in THF (20 +
30 mL) at −85 → −70 °C (30 min) and 0 °C (1 h) provided 1,5-
hexadiene ( )-1d (1.44 g, 12.7 mmol, 78%) after flash chromatog-
raphy (cyclohexane/ethyl acetate 100/1): Rf 0.57 (cyclohexane/ethyl
acetate 5/1); 1H NMR (CDCl3, 500 MHz, δ) 1.27 (d, J = 6.0 Hz, 3H),
1.28 (d, J = 6.1 Hz, 3H), 2.43 (dd, J = 14.0, 6.5 Hz, 1H), 2.58 (dd, J =
14.0, 7.9 Hz, 1H), 3.34 (s, 1H), 5.06 (spt, J = 6.3 Hz, 1H), 5.11 (s,
1H), 5.14 (d, J = 5.0 Hz, 1H), 5.19 (dd, J = 10.6, 1.2 Hz, 1H), 5.50
(dd, J = 17.1, 1.2 Hz, 1H), 5.72−5.83 (m, 1H), 5.98 (dd, J = 17.1, 10.5
Hz, 1H); 13C NMR (CDCl3, 126 MHz, δ) 21.8 (CH3), 21.9 (CH3),
43.6 (CH2), 70.3 (CH), 77.1 (C), 115.3 (CH2), 119.1 (CH2), 132.2
(CH), 138.6 (CH), 174.1 (C); IR (cm−1) 3520(br m) (ν OH),
2985(m), 2960(s), 2940(w), 1725(s) (ν CO), 1645(m) (ν CC),
1470(w), 1435(w), 1375(m), 1275(m), 1225(s) (ν C−O−C ester),
1175(s), 1105(s), 995(m). Anal. Calcd. for C10H16O3: C, 65.2; H, 8.8.
Found: C, 65.1; H, 9.1; C10H16O3, M =184.23 g/mol.
1,5-Hexadiene ( )-1f. According to the general procedure for
the [2,3]-Wittig rearrangement, the reaction of allyl vinyl ether 1146
(2.12 g, 10 mmol) with (i-Pr)2NH (1.69 mL, 12.0 mmol, 1.2 equiv)
and n-BuLi (4.78 mL, 11.0 mmol, 1.1 equiv, 2.3 M in hexanes) in THF
(20 mL + 30 mL) at −85 → −70 °C (30 min) and −15 °C instead of
0 °C (45 min) provided 1,5-hexadiene ( )-1f (1.75 g, 8.3 mmol, 83%)
after flash chromatography (cyclohexane/ethyl acetate 200/1): Rf 0.69
(cyclohexane/ethyl acetate 5/1); 1H NMR (CDCl3, 500 MHz, δ) 1.06
(s, 3H), 1.10 (s, 3H), 1.27 (d, J = 6.3 Hz, 3H), 1.30 (d, J = 6.3 Hz,
3H), 3.43 (s, 1H), 5.01 (dd, J = 13.9, 1.1 Hz, 1H), 5.04 (dd, J = 7.5,
1.1 Hz, 1H), eclipsed by 5.07 (spt, J = 6.2, 1H), and 5.23 (dd, J = 10.6,
1.8 Hz, 1H), 5.51 (dd, J = 17.0, 1.8 Hz, 1H), 6.00 (dd, J = 17.4, 10.9
Hz, 1H), 6.12 (dd, J = 17.0, 10.6 Hz, 1H); 13C NMR (CDCl3, 101
MHz, δ) 21.9 (CH3), 22.0 (CH3), 22.0 (CH3), 22.5 (CH3), 44.3 (C),
70.6 (CH), 80.9 (C), 113.2 (CH2), 116.4 (CH2), 135.7 (CH), 143.8
(CH), 174.2 (C); IR (cm−1) 3505(br m) (ν OH), 2985(s), 2935(m),
1715(s) (ν CO), 1640(w) (ν CC), 1470(m), 1455(w), 1375(m),
1270(s) (ν C−O−C ester), 1245(s), 1180(s), 1150(s), 1105(s),
J
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
995(m). Anal. Calcd. for C12H20O3: C, 67.9; H, 9.5. Found: C, 67.8;
H, 9.5; C12H20O3, M = 212.29 g/mol.
(5-CH), 174.9 (CO), 176.7 (CO); significant COSY cross peaks
1-CH/1′-CH2, 1-CH/5-CH, 5-CH/4-CH, no cross peak observed for
4′-CH2/2′-CH2; significant HMBC cross peaks 1′-CH2/3′-C, 1′-CH2/
2-CH; IR (cm−1) 3515(br m) (ν OH), 2980(m), 2955(s), 2935(m),
2870(m), 1720(s) (ν CO), 1640(w) (ν CC), 1465(m),
1375(m), 1255(s) (ν C−O−C ester), 1230(m), 1205(m), 1145(m),
1105(s), 1175(w); HRMS (ESI) Calcd. for C26H45O6 ([M + H]+)
453.32107, found 453.32133; C26H44O2, M = 452.62 g/mol.
1,5-Hexadiene ( )-1g. According to the general procedure for the
[2,3]-Wittig rearrangement, the reaction of allyl vinyl ether 1242 (1.13
g, 5.7 mmol) with (i-Pr)2NH (0.96 mL, 6.8 mmol, 1.2 equiv) and n-
BuLi (2.73 mL, 6.3 mmol, 1.1 equiv, 2.3 M in hexanes) in THF (11
mL + 17 mL) at −85 → −70 °C (30 min) and 0 °C (1 h) provided
1,5-hexadiene ( )-1g (0.77 g, 3.9 mmol, 69%) after flash
chromatography (n-pentane/diethyl ether 100/1): Rf 0.57 (cyclo-
1
hexane/ethyl acetate 5/1); H NMR (CDCl3, 400 MHz, δ) 1.26 (d,
1
1
1
( )-7b: H,1H COSY, H,13C HSQC, H,13C HMBC and NOESY
experiments were employed to confirm the NMR peak assignments;
1H NMR (CDCl3, 500 MHz, δ) 0.85 (t, J = 7.2 Hz, 6H, 9/9′-CH3),
1.13−1.37 (m, 8H, 7/7′-CH2+8/8′-CH2), eclipsed by 1.24 (d, J = 6.3
Hz, 6H, 2 × i-Pr-CH3), and 1.28 (d, J = 6.3 Hz, 6H, 2 × i-Pr-CH3),
1.46 (dddd, J = 11.4, 11.4, 11.4, 4.2 Hz, 2H, 5/5′-CH2Re), 1.67 (dddd,
J = 6.3 Hz, 3H), 1.29 (d, J = 6.3 Hz, 3H), 1.77 (s, 3H), 2.38 (d, J =
13.9 Hz, 1H), 2.61 (d, J = 13.9 Hz, 1H), 3.35 (s, 1H), 4.71−4.91 (m,
2H), 5.06 (spt, J = 6.3 Hz, 1H), 5.17 (dd, J = 10.5, 1.2 Hz, 1H), 5.50
(dd, J = 17.0, 1.2 Hz, 1H), 5.99 (dd, J = 17.0, 10.5 Hz, 1H); 13C NMR
(CDCl3, 101 MHz, δ) 21.8 (CH3), 21.9 (CH3), 24.2 (CH3), 46.6
(CH2), 70.3 (CH), 77.5 (C), 114.8 (CH2), 115.1 (CH2), 139.2 (CH),
141.2 (C), 174.4 (C); 3515(br m) (ν OH), 2980(m), 2935(m),
1725(s) (ν CO), 1645(w) (ν CC), 1470(w), 1455(m), 1375(m),
1275(m) (ν C−O−C ester), 1210(s), 1145 (s), 1105(s), 1060(m).
Anal. Calcd. for C11H18O3: C, 66.6; H, 9.2. Found: C, 66.7; H, 9.4;
C11H18O3, M = 198.26 g/mol.
Si
J = 11.50, 11.50, 11.50, 5.10 Hz, 2H, 6/6′-CH2 ), 1.81−1.90 (m, 2H,
6/6′-CH2Re), 1.90−2.00 (m, 2H, 5/5′-CH2Re), 2.31 (tt, J = 8.3 Hz, 2H,
4/4′-CH), 2.75−2.84 (m, 2H, 2/2′-CH), 3.08 (s, 2H, 2 × OH), 5.08
(spt, J = 6.3 Hz, 2H, 2 × i-Pr-CH), 5.31 (dd, J = 5.1, 2.4 Hz, 2H, 1/1′-
CH); 13C NMR (CDCl3, 126 MHz, δ) 14.5 (9/9′-CH3), 21.2 (8/8′-
CH2), 21.9 (2 × i-Pr-CH3), 22.1(2 × i-Pr-CH3), 29.0 (6/6′-CH2), 29.3
(5/5′-CH2), 31.6 (7/7′-CH2), 47.8 (4/4′-CH), 53.0 (2/2′-CH), 69.6
(2 × i-Pr-CH), 84.8 (3/3′-C), 131.1 (1/1′-CH), 176.0 (2 × CO);
significant COSY cross peaks 1-CH/2-CH, 2-CH/6-CH2; significant
Si
Re
NOESY cross peaks 1-CH/6-CH2 , 2-CH/4-CH, 2-CH/6-CH2
,
Si
Si
PdII-Catalyzed Cycloisomerization: Representative Proce-
dure for the Kinetic Experiments. [Pd(allyl)(MeCN)2](BF4),47
[Pd(MeCN)4](BF4)2,48 as well as ( )-1b,c were prepared according
to a published procedure.20
4-CH/5-CH2 ; significant HMBC cross peaks OH/4-CH; 6-CH2 /
1-CH; IR (cm−1) 3470(br s) (ν OH), 29(m), 2930(m), 2870(m),
1715(s) (ν CO), 1620(m) (ν CC), 1455(s), 1385(s), 1270(m)
(ν C−O−C ester), 1220(m), 1105(s), 1035(s); HRMS (ESI) Calcd.
for C26H45O6 ([M + H]+) 453.32107, found 453.32135; C26H44O2,
M = 452.62 g/mol.
A solution of hexadiene ( )-syn-1b (4.5 mg, 0.02 mmol, 1 equiv) in
dry CDCl3 (0.2 mL, 10 mL/mmol 1b) was added to an oven-dried
Shigemi tube under argon. The tube was sealed with a septum, and
a
1H NMR spectrum was recorded (substrate spectrum). [Pd-
(MeCN)4](BF4)2 (0.05 mg, 0.001 mmol, 0.05 equiv) was added, and
the reaction was monitored by 1H NMR spectroscopy at hourly
intervals up to 69 h. The degree of conversion was determined in
reference to the substrate spectrum. The NMR resonances could be
assigned to ( )-cis-2b, ( )-cis-3b, 5b, and the dimeric cyclo-
isomerization products 6b and 7b. For NMR kinetic experiments,
1H NMR measurements were performed with a 500 MHz NMR
spectrometer equipped with a 5 mm triple resonance probe H(C,X).
All measurements used identical conditions (spectral width 5500 Hz,
acquisition time 5.96 s, 64 kB data, 90° pulse width 10.3 μs).
The kinetic measurements were performed as array experiments.
In this case, the preacquisition time was arrayed from 0 to 88 h.
Each spectrum was measured with a single scan. The processing was
executed by using zero filling (256 kB).
General Procedure for the PdII-Catalyzed Cycloisomerization.
The corresponding Pd-catalyst (0.05 equiv) was added to a solu-
tion of 1,5-hexadiene ( )-1 (1 equiv) in dry CH2Cl2 (10 mL/mmol 1),
and the resulting mixture was stirred at the given temperature (reaction
at elevated temperatures were carried out in a glass pressure tube,
sealed with a Teflon screwcap) until TLC indicated complete
conversion of the starting material (3−23 h). The reaction mixture was
filtered through a short silica gel-plug (eluent CH2Cl2 or ethyl acetate)
to remove the catalyst and was then concentrated under reduced
pressure. Purification by flash chromatography (cyclohexane/ethyl
acetate) afforded the corresponding methylene cyclopentanes ( )-2/2′
along with methyl cyclopentene ( )-3 and the mono- and double
cycloisomerized dimers ( )-6 and ( )-7 as a mixture of regioisomers
and diastereomers.
Dimer ( )-6b: Rf 0.51 (cyclohexane/ethyl acetate 5/1); the NMR
peak assignment is based on 1H,1H COSY, 1H,13C HSQC, and 1H,13C
1
HMBC experiments; H NMR (CDCl3, 500 MHz, δ) 0.84 (t, J = 7.2
Cycloalkenes ( )-2c and ( )-6c. According to the general
procedure for the PdII-catalyzed cycloisomerization, 1,5-hexadiene
( )-1c (110 mg, 0.4 mmol, syn/anti = 16/84) in CH2Cl2 (4 mL) was
treated with [Pd(MeCN)4](BF4)2 (8.8 mg, 0.02 mmol, 0.05 equiv) at
ambient temperature for 16 h. Purification by flash chromatography
(cyclohexane/ethyl acetate 100/1 → 50/1) afforded methylene
cyclopentane ( )-2c20 (18 mg, 0.05 mmol, 14%) as a mixture of
diastereomers (cis/trans = 6/94) along with mono cycloisomerized
dimer ( )-6c (25 mg, 0.09 mmol, 23%) as a mixture of diastereomers
(dr = 26/74).
( )-6c: Rf 0.14 (cyclohexane/ethyl acetate 5/1); the NMR peak
assignment is based on 1H,1H COSY and 1H,13C HSQC experiments;
1H NMR (CDCl3, 500 MHz, mixture of diastereomers dr = 26/74, δ)
1.16−1.29 (m, 1H, 5′-CH2), 1.40−1.51 (m, 1H, 6′-CH2), 1.68−1.81
(m, 1H, 6′-CH2), 1.89−1.98 (m, 1H, 5′-CH2), 2.03−2.14 (m, 1H,
Hz, 6H, 9-CH3+9′CH3), 1.02−1.22 (m, 4H), 1.23 (d, J = 6.3 Hz, 3H,
i-Pr-CH3), eclipsed by 1.22−1.30 (m, 2H) and 1.26 (d, J = 6.3 Hz, 3H,
i-Pr-CH3), and 1.29 (d, J = 6.2 Hz, 3H, i-Pr-CH3), and 1.29 (d, J = 6.2
Hz, 3H, i-Pr-CH3), 1.33−1.46 (m, 4H), 1.82−1.93 (m, 2H, 5′/6′-
CH2), 2.00−2.12 (m, 2H, 1′-CH2), 2.16−2.27 (m, 1H, 4′-CH), 2.27−
2.39 (m, 2H, 4-CH+2′-CH), 2.99 (s, 1H, OH), 3.26 (s, 1H, OH), 4.97
(dd, J = 17.2, 1.7 Hz, 1H, 6-CH2), 5.04 (spt, J = 6.2 Hz, 1H, i-Pr-CH),
eclipsed by 5.06−5.14 (m, 1H, 6-CH2), and 5.10 (spt, J = 6.2 Hz, 1H,
i-Pr-CH), 5.46 (d, J = 15.3 Hz, 1H, 2-CH), 5.50 (ddd, J = 17.3, 9.9, 9.9
Hz, 1H, 5-CH), 5.66 (dt, J = 15.3, 7.3 Hz, 1H, 1-CH); 13C NMR
(CDCl3, 126 MHz, δ) 14.0 (9/9′-CH3), 14.5 (9/9′-CH3), 20.4 (CH2),
21.1, (CH2), 21.9/21.9/21.9/21.9 (4 × i-Pr-CH3), 28.8 (CH2), 28.8
(CH2), 31.0 (CH2), 31.3 (CH2), 32.2 (1′-CH2), 48.4 (2′-CH), 48.7
(4′-CH), 51.3 (4-CH), 69.5 (i-Pr-CH), 70.2 (i-Pr-CH), 79.1 (3-C),
83.4 (3′-C), 117.9 (6-CH2), 129.2 (1-CH), 131.4 (2-CH), 137.0
K
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
1′-CH2), 2.16−2.23 (m, 1H, 1′-CH2), 2.23−2.32 (m, 1H, 2′-CH),
2.40−2.54 (m, 1H, 4′-CH), 2.73 (ddd, J = 9.1, 7.0, 4.2 Hz, 1H, 4-CH),
3.23 (s, 0.7Hmajor, OH), 3.24 (s, 0.3Hminor, OH), 3.32−3.43 (m, 2H,
7′-CH2), 3.53−3.70 (m, 6H, 7-CH2+OH+OMe), 3.68−3.76 (m, 3H,
OMe), 4.35−4.54 (m, 4H, 2 × OCH2Bn), 5.11−5.19 (m, 2H, 6-CH2),
( )-6d/( )-7d: Rf 0.11 (cyclohexane/ethyl acetate 5/1); the NMR
peak assignment is based on 1H,1H COSY and 1H,13C HSQC
1
experiments; H NMR (CDCl3, 500 MHz, mixture of diastereomers
and regioisomers, 6d/7d = 73/27, dr6d = 74/26, δ) (for clarity, only
selected resonances of 7d are listed) 1.21−1.31 (m, 12H, i-Pr-CH3),
1.39−1.53 (m, 1H, CH2), 1.62−1.96 (m, 4H, 2 × CH2), 2.02−2.27
(m, 4H, 1′-CH2+2′-CH+CH2), 2.33−2.43 (m, 1H, 4-CH2), 2.51 (dd,
J = 14.0, 7.5 Hz, 1H, 4-CH2), 2.61−2.71 (m, 0.2Hminor, 2/2′-CH), 3.11
(s, 0.8Hmajor, OH), 3.13 (s, 0.2Hminor, OH), 3.20 (s, 0.2Hminor, OH),
3.28 (s, 0.2Hminor, OH), 3.30 (s, 0.8Hmajor, OH), 5.05 (spt, J = 6.2 Hz,
2H, i-Pr-CH), 5.09 (d, J = 12.6 Hz, 2H, 6-CH2), 5.28−5.34 (m, 0.2H,
1/1′-CH), 5.58 (d, J = 15.3 Hz, 0.8Hmajor, 2-CH), 5.63 (d, J = 15.3 Hz,
0.2Hminor, 2-CH), 5.68−5.87 (m, 2H, 5-CH+1-CH); 13C NMR
(CDCl3, 126 MHz, mixture of diastereomers and regioisomers, 6d/
7d = 73/27, dr6d = 74/26, δ) 21.8/21.9/21.9/21.9 (i-Pr-CH3), 22.6/
23.2 (CH2), 30.4/30.5/30.6 (CH2), 31.7/31.8 (1′-CH2), 37.7 (CH2),
39.6/39.6 (CH2), 43.8/44.0 (CH2), 48.5/48.7 (2′-CH), 53.1 (2/
2′-CH), 69.6/69.7 (i-Pr-CH), 70.0/70.2 (i-Pr-CH), 76.4, 82.1/83.7
(3/3′-C), 119.0/119.0 (6-CH2), 129.6/129.7 (1-CH), 131.2/131.4
(2-CH), 131.6 (1/1′-CH), 132.4/132.5 (5-CH), 174.4/174.5 (CO),
176.1 (CO), 176.8/176.9 (CO); significant COSY cross peaks
1-CH/1′-CH2; 5-CH/4-CH2, 1-CH/2-CH; IR (cm−1) 3505(br s)
(ν OH), 2980(s), 2940(s), 2875(m), 1725(s) (ν CO), 1640(w)
(ν CC), 1470(m), 1455(m), 1375(s), 1230(s) (ν C−O−C ester),
1180(s), 1145(s), 1110(s), 1035(m); HRMS (ESI) Calcd. for C20H33O6
([M + H]+) 369.22717, found 369.22835; C20H32O6, M = 368.46 g/mol.
5.58 (d, J = 15.3 Hz, 0.7Hmajor, 2-CH), 5.59 (d, J = 15.3 Hz, 0.3Hminor
,
2-CH), 5.79 (dt, J = 15.3, 7.3 Hz, 1H, 1-CH), 5.83−5.92 (m, 1H,
5-CH), 7.23−7.35 (m, 10H, HAr); 13C NMR (CDCl3, 126 MHz,
mixture of diastereomers dr = 26/74, δ) 27.1/27.1 (5′-CH2), 29.9/30.1
(6′-CH2), 31.7/31.8 (1′-CH2), 47.5/47.7 (2′-CH), 51.5/51.6 (4-CH),
52.7 (OMe), 52.9/53.0 (OMe), 53.6/53.6 (4′-CH), 69.9/70.0 (7-
CH2), 71.3 (7′-CH2), 73.0/73.1 (OCH2Bn), 73.4/73.4 (OCH2Bn),
79.3 (3/3′-C), 82.9/83.0 (3/3′-C), 118.8/118.8 (6-CH2), 127.6/127.6
(2 × −CHAr), 127.7/127.8 (4 × −CHAr), 128.5 (4 × −CHAr), 130.4/
130.5 (2-CH+1-CH), 135.7/135.7 (5-CH), 138.2/138.3 (2 × CAr),
174.5 (CO), 176.2 (CO); significant COSY cross peaks 5-CH/
4-CH, 1-CH/1′-CH2, 7-CH2/4-CH, 7′-CH2/4′-CH; IR (cm−1
)
3510(br w) (ν OH), 2955(m), 2865(m), 1730(s) (ν CO),
1495(w), 1455(m), 1435(w), 1385(m), 1365(w), 1255(m) (ν
C−O−C ester), 1095(m), 1030(w), 910(s); HRMS (ESI) Calcd.
for C32H41O8 ([M + H]+) 553.27959, found 553.28026; C32H40O8,
M = 552.66 g/mol.
Cycloalkenes ( )-2d, ( )-3d, ( )-6d, and ( )-7d. According to
the general procedure for the PdII-catalyzed cycloisomerization,
1,5-hexadiene ( )-1d (184 mg, 1 mmol) in dry CH2Cl2 (10 mL)
was treated with [Pd(MeCN)4](BF4)2 (22.0 mg, 0.05 mmol,
0.05 equiv) at ambient temperature for 23 h. Purification by flash
chromatography (cyclohexane/ethyl acetate 100/1 → 5/1) afforded
methylene cyclopentane ( )-2d (15 mg, 0.08 mmol, 8%) along with a
mixed fraction of ( )-2d and ( )-3d (14 mg, 0.08 mmol, 8%, 2d/3d =
72/28) and a mixed fraction consisting of the dimers ( )-6d and ( )-7d
(58 mg, 0.16 mmol, 32%, 6d/7d = 73/27; dr6d = 74/26) as a mixture of
regioisomers and diastereomers.
Cycloalkenes ( )-2e, ( )-2′e, ( )-3e, and ( )-7e. According to
the general procedure for the PdII-catalyzed cycloisomerization, 1,5-
hexadiene ( )-1e (100 mg, 0.5 mmol, syn/anti = 77/23) in dry
CH2Cl2 (5 mL) was treated with [Pd(MeCN)4](BF4)2 (11.2 mg,
0.05 mmol, 0.05 equiv) at ambient temperature for 3 h. Purification by
flash chromatography (n-pentane/diethyl ether 50/1 → 5/1) afforded
cycloalkenes ( )-2e, ( )-2′e, and ( )-3e (11 mg, 0.06 mmol, 11%) as a
mixture of regioisomers and diastereomers (2e/2′e/3e = 61/14/24)
along with ( )-7e (37 mg, 0.1 mmol, 37%). Subsequent recrystallization
of ( )-7e by vapor diffusion technique from isohexane and ethyl acetate
provided colorless crystals suitable for an X-ray crystal structure analysis.
( )-2e, ( )-2′e, and ( )-3e: Rf 0.37 (cyclohexane/ethyl acetate
5/1); the NMR peak assignment is based on 1H,1H COSY and 1H,13C
( )-2d: Rf 0.28 (cyclohexane/ethyl acetate 5/1); the NMR peak
assignment is based on 1H,1H COSY and 1H,13C HSQC experiments;
1H NMR (CDCl3, 500 MHz, δ) 1.27 (d, J = 6.2 Hz, 6H, i-Pr-CH3),
1.89 (t, J = 10.0 Hz, 1H, 2-CH2), 2.14 (dt, J = 12.6, 9.9 Hz, 1H,
2-CH2), 2.41−2.54 (m, 2H, 3-CH2+5-CH2), 2.56−2.66 (m, 1H,
3-CH2), 2.83 (d, J = 16.4 Hz, 1H, 5-CH2), 3.13 (br s, 1H, OH), 4.93
(d, J = 13.7 Hz, 2H, 4′-CH2), 5.09 (spt, J = 6.2 Hz, 1H, i-Pr-CH); 13
C
NMR (CDCl3, 126 MHz, δ) 21.8 (2 × i-Pr-CH3), 31.1 (3-CH2), 38.7
(2-CH2), 45.7 (5-CH2), 69.8 (i-Pr-CH), 81.1 (1-C), 107.1 (4′-CH2),
149.3 (4-C), 175.9 (CO); significant COSY cross peaks 4′-CH2/
5-CH2; IR (cm−1) 3515(br s) (ν OH), 2985(s), 2940(m), 1725(s)
(ν CO), 1645(m) (ν CC), 1670(m), 1435(m), 1390(m),
1375(s), 1280(s) (ν C−O−C Ester), 1230(s), 1175(s), 1105(s)
(ν C−O−C Ether), 995(s); HRMS (ESI) Calcd. for C10H17O3 ([M +
H]+) 185.11722, found 185.11720; C10H16O3, M = 184.23 g/mol.
1
HSQC experiments; H NMR (CDCl3, 500 MHz, mixture of regio-
and diastereomers, 2e/2′e/3e = 61/14/24, δ) 0.88 (t, J = 7.1 Hz,
3H2′e+3e), 0.94−1.05 (m, 3H2e), 1.18−1.32 (m, 6H), 1.68 (br s, 3H3e),
1.87 (ddd, J = 13.0, 7.5, 3.6 Hz, 1H2e), 2.17 (dt, J = 13.1, 10.3 Hz,
1H2e), 2.32−2.64 (m, 2H), 2.71−2.89 (m, 1H), 2.93 (dd, J = 16.6, 2.1
Hz, 1H3e), 3.01 (s, 1H2e), 3.13 (s, 1H3e), 3.20 (s, 1H2′e), 3.48 (q, J =
6.9 Hz, 1H), 4.77 (d, J = 2.3 Hz, 0.2Hminder,2e), 4.82 (d, J = 2.7 Hz,
0.8Hhaupt,2′e), 4.84 (d, J = 2.3 Hz, 0.2Hminder,2e), 4.96 (d, J = 2.3 Hz,
0.8Hhaupt,2e), 5.00 (t, J = 2.3 Hz, 1H2′e), 5.03 (t, J = 2.3 Hz, 1H2′e),
5.07−5.15 (m, 1H), 5.33 (s, 1H3e); IR (cm−1) 3510(br s) (ν OH),
2980(s), 2940(s), 2875(m), 1725(s) (ν CO), 1640(s) (ν CC),
1470(m), 1455(m), 1455(s), 1400(m), 1375(s), 1260(s) (ν C−O−C
ester), 1165(s), 1145(s), 1105(s) (ν C−O−C Ether), 995(s); HRMS
(ESI) Calcd. for C11H19O3 ([M + H]+) 199.13287, found 199.13277;
C11H18O3, M = 198.26 g/mol.
( )-2d/( )-3d: Rf 0.26 (cyclohexane/ethyl acetate 5/1); 1H
NMR (CDCl3, 400 MHz, mixture of regioisomers, 2d/3d = 72/28, δ)
1.21−1.33 (m, 6H), 1.79−2.32 (m, 2H), 2.35−2.71 (m, 2H),
2.75−2.98 (m, 1.4H2d+0.3H3d), 3.14 (s, 0.7H2d), 3.35 (s, 0.3H3d),
4.88−4.98 (m, 1.4H2d), 5.01−5.16 (m, 1H), 5.28 (s, 0.3H3d);
M = 184.23 g/mol.
( )-cis-7e: Rf 0.20 (cyclohexane/ethyl acetate 5/1); the NMR peak
assignment is based on 1H,1H COSY and 1H,13C HSQC experiments;
L
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
1H NMR (CDCl3, 500 MHz, δ) 0.88 (d, J = 6.9 Hz, 6H, 7/7′-CH3),
1.24 (d, J = 6.2 Hz, 6H, i-Pr-CH3), 1.25 (d, J = 6.2 Hz, 6H, i-Pr-CH3),
1.45 (ddt, J = 11.6, 11.5, 4.2 Hz, 2H, 5/5′-CH2), 1.65 (ddt, J = 12.2,
11.1, 5.0 Hz, 2H, 6/6′-CH2), 1.79−1.87 (m, 2H, 6/6′-CH2), 1.87−1.94
(m, 2H, 5/5′-CH2), 2.36 (tq, J = 9.6, 6.9 Hz, 2H, 4/4′-CH), 2.85
(dddd, J = 10.5, 8.0, 5.6, 2.3 Hz, 2H, 2/2′-CH), 2.92 (s, 2H, OH), 5.07
(spt, J = 6.2 Hz, 2H, i-Pr-H), 5.41 (dd, J = 5.2, 2.5 Hz, 2H, 1/1′-CH);
13C NMR (CDCl3, 126 MHz, δ) 12.9 (7/7′-CH3), 22.0 (2 × i-Pr-
CH3), 22.1 (2 × i-Pr-CH3), 29.2 (6/6′-CH2), 30.6 (5/5′-CH2), 43.0
(4/4′-CH), 52.7 (2/2′-CH), 69.5 (2 × i-Pr-CH), 85.1 (3/3′-C), 130.8
(1/1′-CH), 175.5 (2 × CO); significant COSY cross peaks 1-CH/
2-CH, 2-CH/6-CH2, 4-CH/7-CH3, C22H36O6, M = 396.52 g/mol.
1145(s), 1110(s). Anal. Calcd. for C12H20O3: C, 67.9; H, 9.5. Found:
C, 67.9; H, 9.6; C12H20O3, M = 212.29 g/mol.
( )-3f: Rf 0.49 (cyclohexane/ethyl acetate 5/1); the NMR peak
assignment is based on 1H,1H COSY, 1H,13C HSQC, and 1H,13C
1
HMBC experiments; H NMR (CDCl3, 500 MHz, δ) 0.90 (s, 3H,
7-CH3), 1.08 (s, 3H, 7-CH3), 1.28 (d, J = 6.3 Hz, 3H, i-Pr-CH3),
eclipsed by 1.28 (d, J = 6.3 Hz, 3H, i-Pr-CH3), 1.61 (d, J = 1.7 Hz, 3H,
6-CH3), 2.35 (dt, J = 16.6, 2.0 Hz, 1H, 2-CH2), 3.02 (dt, J = 16.6, 2.1
Hz, 1H, 2-CH2), eclipsed by 3.03 (s, 1H, OH), 5.09 (spt, J = 6.3 Hz,
1H, i-Pr-CH), 5.29 (br s, 1H, 1-CH); 13C NMR (CDCl3, 126 MHz, δ)
12.8 (6-CH3), 19.4 (7-CH3), 21.9 (i-Pr-CH3), 22.0 (i-Pr-CH3), 22.6
(7-CH3), 40.5 (2-CH2), 52.8 (4-C), 69.4 (i-Pr-CH), 86.1 (3-C), 119.9
(1-CH), 144.5 (5-C), 174.5 (CO); significant COSY cross peaks
1-CH/6-CH3; significant HMBC cross peaks 1-CH/4-C, 1-CH/3-C,
OH/2-CH2; IR (cm−1) 3500(br s) (ν OH), 2980(s), 2935(s),
2875(s), 1765(m) (ν CC), 1725(s) (ν CO), 1465(s), 1385(s),
1375(s), 1265(s) (ν C−O−C ester), 1205(s), 1185(s), 1145(s),
1110(s), 1055(s), 1020(s); Calcd. for C12H21O3 ([M + H]+)
213.14852, found 213.14818; C12H20O3, M = 212.29 g/mol.
Cycloalkenes ( )-2f, ( )-2′f, ( )-3f, ( )-6f, and ( )-7f. According
to the general procedure procedure for the PdII-catalyzed cyclo-
isomerization, 1,5-hexadiene ( )-1f (100 mg, 0.47 mmol) in dry
CH2Cl2 (4.7 mL) was treated with [Pd(MeCN)4](BF4)2 (10.4 mg,
0.024 mmol, 0.05 equiv) and PCy3 (6.7 mg, 0.024 mmol, 0.05 equiv),
at room temperature for 6 h. Purification by flash chromatography
(cyclohexane/ethyl acetate 100/1 → 5/1) afforded regioisomeric
methylene cyclopentane ( )-2f′ (6 mg, 0.03 mmol, 6%) along with a
mixed fraction of cycloalkenes ( )-2f and ( )-3f (21 mg, 0.1 mmol,
21%, 2f/3f = 89/11) and a dimer fraction containing ( )-6f and
( )-7f (52 mg, 0.13 mmol, 52%) as mixture of regioisomers and
diastereomers (6f/7f = 60/40, dr7f = 66/34). Subsequent recrystalliza-
tion by vapor diffusion technique from isohexane and ethyl acetate
provided colorless crystals of ( )-cis-7f suitable for an X-ray crystal
structure analysis.
( )-2′f: Rf 0.77 (cyclohexane/ethyl acetate 5/1); the NMR peak
assignment is based on 1H,1H COSY, 1H,13C HSQC, and 1H,13C
HMBC experiments; 1H NMR (CDCl3, 500 MHz, δ) 0.93 (s, 6H, 2 ×
7-CH3), 1.23 (d, J = 6.3 Hz, 3H, i-Pr-CH3), 1.30 (d, J = 6.3 Hz, 3H,
i-Pr-CH3), 1.54 (ddd, J = 12.4, 6.9, 5.2 Hz, 1H, 5-CH2), 1.94 (dt, J =
12.4, 9.7 Hz, 1H, 5-CH2), 2.47−2.53 (m, 2H, 6-CH2), 3.72 (s, 1H,
OH), 5.03 (t, J = 2.0 Hz, 1H, 1-CH2), 5.07 (spt, J = 6.3 Hz, 1H, i-Pr-
CH), 5.15 (t, J = 2.51 Hz, 1H, 1-CH2); 13C NMR (CDCl3, 126 MHz, δ)
21.7 (i-Pr-CH3), 21.9 (i-Pr-CH3), 22.6 (7-CH3), 24.1 (7-CH3), 27.3
(6-CH2), 35.6 (6-CH2), 44.8 (4-C), 70.2 (i-Pr-CH), 85.9 (3-C), 109.2
(1-CH2), 152.5 (2-C), 175.0 (CO); significant COSY cross peaks
1-CH2/6-CH2; significant HMBC cross peaks 1-CH2/6-C, 1-CH2/3-C;
IR (cm−1) 3480(br m) (ν OH), 2960(s), 2930(s), 2870(s), 1725(s)
(ν CO), 1635(w) (ν CC), 1465(s), 1385(s), 1375(s), 1270(s)
(ν C−O−C ester), 1235(s), 1180(s), 1145(s), 1110(s), 1050(s);
HRMS (ESI) Calcd. for C12H20O3Na ([M + Na]+) 235.13047, found
235.13027; C12H20O3, M = 212.29 g/mol.
( )-6f: Rf 0.40 (cyclohexane/ethyl acetate 5/1); the NMR peak
assignment is based on 1H,1H COSY and 1H,13C HSQC experiments;
1H NMR (CDCl3, 500 MHz, δ) 0.96 (s, 3H, 7/7′-CH3), 0.98 (s, 3H,
7/7′-CH3), 1.03 (s, 3H, 7/7′-CH3), 1.07 (s, 3H, 7/7′-CH3), 1.26 (d,
J = 6.3 Hz, 3H, i-Pr-CH3), eclipsed by 1.26−1.28 (m, 6H, 2 × i-Pr-
CH3), and 1.29 (d, J = 6.3 Hz, 3H, i-Pr-CH3), 1.37−1.42 (m, 1H,
6′-CH2), 1.46 (dt, J = 8.7, 4.4 Hz, 1H, 5′-CH2), 1.73−1.82 (m, 1H,
5′-CH2), 1.85−1.95 (m, 1H, 6′-CH2), 2.08−2.22 (m, 2H, 1′-CH2), 2.76
(dddd, J = 9.0, 9.0, 9.0, 6.0 Hz, 1H, 2′-CH), 3.03 (s, 1H, OH), 3.41 (s,
1H, OH), 4.98 (dd, J = 16.4, 1.3 Hz, 1H, 6-CH2), eclipsed by 5.01−
5.03 (m, 1H, 6-CH2), and 5.05 (spt, J = 6.3 Hz, 1H, i-Pr-CH), and
5.07 (spt, J = 6.2 Hz, 1H, i-Pr-CH3), 5.73 (d, J = 15.3 Hz, 1H, 2-CH),
5.82 (dt, J = 15.3, 7.3 Hz, 1H, 1-CH), 5.97 (dd, J = 17.5, 10.9 Hz, 1H,
5-CH); 13C NMR (CDCl3, 126 MHz, δ) 21.9/22.0/22.0/22.1 (4 ×
i-Pr-CH3), 22.1/22.4/22.6/26.7 (4 × 7/7′-CH3), 27.6 (6′-CH2), 33.5
(1′-CH2), 38.8 (5′-CH2), 44.1 (2′-CH), 44.5/47.8 (4-C+4′-C), 69.6
(i-Pr-CH), 70.5 (i-Pr-CH), 80.4/86.2 (3-C+3′-C), 113.0 (6-CH2),
128.3 (2-CH), 131.1 (1-CH), 144.1 (5-CH), 174.6 (CO), 174.8
(CO); significant COSY cross peaks 5-CH/6-CH, 1-CH/1′-CH2,
2′-CH/1′-CH2, 2′-CH/6′-CH2; IR (cm−1) 3510(br s) (ν OH), 2980(s),
2940(s), 2875(s), 1715(s) (ν CO), 1640(w) (ν CC), 1470(s),
1385(s), 1375(s), 1270(s) (ν C−O−C ester), 1225(s), 1180(m),
1145(s), 1110(s); Calcd. for C24H41O6 ([M + H]+) 425.28977, found
425.28963; C24H40O6, M = 424.57 g/mol.
( )-2f: Rf 0.49 (cyclohexane/ethyl acetate 5/1); the NMR peak
assignment is based on 1H,1H COSY, 1H,13C HSQC, and 1H,13C
1
HMBC experiments; H NMR (CDCl3, 500 MHz, δ) 0.95 (s, 3H,
7-CH3), 1.09 (s, 3H, 7-CH3), 1.25 (d, J = 6.3 Hz, 3H, i-Pr-CH3), 1.27
(d, J = 6.3 Hz, 3H, i-Pr-CH3), 1.92 (ddd, J = 13.6, 10.0, 6.0 Hz, 1H,
2-CH2), 2.23 (ddd, J = 13.6, 9.6, 6.7 Hz, 1H, 2-CH2), 2.47−2.64 (m,
2H, 1-CH2), 3.19 (s, 1H, OH), 4.78 (t, J = 2.4 Hz, 1H, 6-CH2), 4.81
(t, J = 2.0 Hz, 1H, 6-CH2), 5.06 (spt, J = 6.3 Hz, 1H, i-Pr-CH); 13C
NMR (CDCl3, 126 MHz, δ) 21.9 (i-Pr-CH3), 22.0 (i-Pr-CH3), 23.1
(7-CH3), 24.1 (7-CH3), 28.5 (1-CH2), 32.3 (2-CH2), 49.5 (4-C), 69.8
(i-Pr-CH), 85.4 (3-C), 104.4 (6-CH2), 158.9 (5-C), 175.1 (CO);
significant COSY cross peaks 6-CH2/1-CH2; significant HMBC cross
peaks 6-CH2/4-C, 6-CH2/1-CH2, OH/2-CH2; IR (cm−1) 3510(br s)
(ν OH), 2980(s), 2935(s), 2875(s), 1715(s) (ν CO), 1655(m) (ν
CC), 1465(s), 1375(s), 1255(s) (ν C−O−C ester), 1195(s),
( )-7f: Rf 0.40 (cyclohexane/ethyl acetate 5/1); the NMR peak
assignment is based on 1H,1H COSY, 1H,13C HSQC, and 1H,13C
1
HMBC experiments; H NMR (CDCl3, 500 MHz, δ) 0.99 (s, 12H,
2 × 7-CH3+2 × 7′-CH3), 1.26 (d, J = 6.3 Hz, 12H, 4 × i-Pr-CH3), 1.49
(td, J = 8.2, 4.0 Hz, 2H, 5/5′-CH2), 1.59−1.68 (m, 2H, 6/6′-CH2),
1.76−1.84 (m, 2H, 5/5′-CH2), 1.85−1.94 (m, 2H, 6/6′-CH2), 2.99 (s,
2H, OH), 3.19−3.28 (m, 2H, 2/2′-CH), 5.03 (spt, J = 6.3 Hz, 2H,
i-Pr-CH), 5.49 (dd, J = 4.9, 2.3 Hz, 2H, 1/1′-CH); 13C NMR (CDCl3,
126 MHz, δ) 22.1 (4 × i-Pr-CH3), 22.6 (7/7′-CH3), 27.2 (7/7′-CH3),
27.8 (6/6′-CH2), 39.0 (5/5′-CH2), 47.6 (2/2′-CH), 48.5 (4/4′-C), 69.7
(2 × i-Pr-CH), 87.4 (3/3′-C), 131.2 (1/1′-CH), 174.4 (2 × CO);
M
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
significant COSY cross peaks 1-CH/2-CH, 2-CH/6-CH2; significant
HMBC cross peaks 1′-CH/2-CH, OH/2-CH; IR (cm−1) 3525(br s)
(ν OH), 2980(s), 2960(s), 2870(s), 1710(s) (ν CO), 1620(w)
(ν CC), 1470(s), 1385(m, 1375(m), 1280(s) (ν C−O−C ester),
1230(s), 1180(w), 1145(w), 1105(s); Calcd. for C24H41O6 ([M + H]+)
425.28977, found 425.28993; C24H40O6, M = 424.57 g/mol.
ASSOCIATED CONTENT
* Supporting Information
Computational details, copies of NMR as well as IR spectra,
and CIF files for compounds 7e,f. This material is available free
of charge via the Internet at http://pubs.acs.org.
■
S
AUTHOR INFORMATION
Corresponding Author
*E-mail: bjoern.nelson@tu-dortmund.de; martin.hiersemann@
tu-dortmund.de.
■
( )-cis-7f: Rf 0.29 (cyclohexane/ethyl acetate 5/1); the NMR peak
assignment is based on 1H,1H COSY and 1H,13C HSQC experiments;
1H NMR (CDCl3, 500 MHz, δ) 0.99 (s, 6H, 2 × 7/7′-CH3), 1.00 (s,
6H, 2 × 7/7′-CH3), 1.25 (d, J = 6.3 Hz, 6H, 2 × i-Pr-CH3), 1.27 (d, J =
6.3 Hz, 6H, 2 × i-Pr-CH3), 1.46−1.53 (m, 2H, 5/5′-CH2), 1.55−1.65
(m, 2H, 6/6′-CH2), 1.76−1.90 (m, 4H, 5/5′-CH2+6/6′-CH2), 3.00
(s, 2H, OH), 3.19−3.30 (m, 2H, 2/2′-CH), 5.07 (spt, J = 6.3 Hz, 2H,
i-Pr-CH), 5.50 (dd, J = 5.2, 2.5 Hz, 2H, 1/1′-CH); 13C NMR (CDCl3,
126 MHz, δ) 22.1 (2 × i-Pr-CH3), 22.2 (2 × i-Pr-CH3), 22.7 (2 × 7/
7′-CH3), 27.2 (2 × 7/7′-CH3), 27.9 (6/6′-CH2), 39.0 (5/5′-CH2), 47.5
(2/2′-CH), 48.6 (4/4′-CH), 69.5 (i-Pr-CH), 87.4 (3/3′-C), 131.3 (1/
1′-CH), 174.4 (2 × CO); significant COSY cross peaks 1-CH/2-
CH, 2-CH/6-CH2; IR (cm−1) 3505(br s) (ν OH), 2980(s), 2975(s),
1715(s) (ν CO), 1640(m) (ν CC), 1470(s), 1415(m), 1385(s),
1270(s) (ν C−O−C ester), 1180(s), 1145(s), 1105(s), 1045(s),
980(s), 910(s); Calcd. for C24H41O6 ([M + H]+) 425.28977, found
425.28956; C24H40O6, M = 424.57 g/mol.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We are grateful to the Deutsche Forschungsgemeinschaft
(HI628/10-1) and the Technische Universitat Dortmund for
■
̈
financial support of this research. Computations were
performed on the LiDOng cluster at the ITMC of the TU
Dortmund.
REFERENCES
■
(1) (a) Trost, B. M. Chem. Soc. Rev 1982, 11, 141−170. (b) Hudlicky,
T.; Price, J. D. Chem. Rev. 1989, 89, 1467−1486. (c) Lautens, M.;
Klute, W.; Tam, W. Chem. Rev. 1996, 96, 49−92. (d) Heasley, B. Eur.
J. Org. Chem. 2009, 1477−1489.
(2) For selected reviews on 1,n-enyne, -diyne, and -diene
cycloisomerization, see: (a) Trost, B. M. Acc. Chem. Res. 1990, 23,
34−42. (b) Trost, B. M.; Krische, M. J. Synlett 1998, 1−16.
(c) Widenhoefer, R. A. Acc. Chem. Res. 2002, 35, 905−913. (d) Aubert,
C.; Buisine, O.; Malacria, M. Chem. Rev. 2002, 102, 813−834.
(e) Lloyd-Jones, G. C. Org. Biomol. Chem. 2003, 1, 215−236.
(f) Schmidt, B. Angew. Chem., Int. Ed. Engl. 2003, 42, 4996−4999.
(g) Ma, S.; Yu, S.; Gu, Z. Angew. Chem., Int. Ed. Engl. 2006, 45, 200−
203. (h) Zhang, L.; Sun, J.; Kozmin, S. A. Adv. Synth. Catal. 2006, 348,
2271−2296. (i) Michelet, V.; Toullec, P. Y.; Genet, J.-P. Angew. Chem.,
Int. Ed. Engl. 2008, 47, 4268−4315.
Dimer ( )-8. According to the general procedure for the PdII-
catalyzed cycloisomerization, 1,5-hexadiene ( )-1g (100 mg, 0.5
mmol) in dry CH2Cl2 (5 mL) was treated with [Pd(MeCN)4](BF4)2
(11.2 mg, 0.025 mmol, 0.05 equiv) at ambient temperature for 3 days.
Purification by flash chromatography (cyclohexane/ethyl acetate 100/
1 → 50/1) afforded ( )-8 (50 mg, 0.13 mmol, 50%) as a mixture of
diastereomers (dr = 71/29): Rf 0.2 (cyclohexane/ethyl acetate 5/1);
the NMR peak assignment is based on 1H,1H COSY and 1H,13C
HSQC experiments; 1H NMR (CDCl3, 500 MHz, mixture of
diastereomers, dr = 71/29, δ) 1.01 (s, 0.9Hminor, 7′-CH3), 1.10 (s,
2.1Hmajor, 7′-CH3), 1.20−1.31 (m, 12H, 4 × i-Pr-CH3), 1.35−1.70 (m,
2H, 4′-CH2), 1.50−2.24 (m, 6H, 6′-CH2+1′-CH2+2′-CH2), 1.76 (s,
3H, 7-CH3), 2.39 (d, J = 13.9 Hz, 1H, 4-CH2), 2.59 (d, J = 13.9 Hz,
1H, 4-CH2), 3.18 (br s, 1H, OH), 3.35 (br s, 1H, OH), 4.77 (s, 1H,
6-CH2), 4.85 (s, 1H, 6-CH2), 5.05 (spt, J = 6.3 Hz, 1H, i-Pr-CH),
eclipsed by 5.06 (spt, J = 6.3 Hz, 1H, i-Pr-CH), 5.62 (d, J = 15.3 Hz,
0.7Hmajor, 2-CH), 5.63 (d, J = 14.9 Hz, 0.3Hminor, 2-CH), 5.85−5.96
(m, 1H, 1-CH); 13C NMR (CDCl3, 126 MHz, mixture of
diastereomers, dr = 71/29, δ) 21.8/21.8 (2 × i-Pr-CH3), 21.9/21.9
(2 × i-Pr-CH3), 24.2 (7-CH3), 27.1/27.2 (7′-CH3), 38.6/38.7 (4′-
CH2), 38.8/39.0 (1′/2′-CH2), 42.9/43.0 (5′-C), 45.0 (6′-CH2), 46.9/
46.9 (4-CH2), 51.8/51.8 (1′/2′-CH2), 69.6/69.6 (i-Pr-CH), 70.1/70.1
(i-Pr-CH), 82.2/82.3 (3-C+3′-C), 114.9/115.0 (6-CH2), 128.0/128.2
(1-CH), 133.7/133.8 (2-CH), 141.3/141.4 (5-C), 174.7/174.7 (C
O), 176.9/176.9 (CO); significant COSY cross peaks 1-CH/6′-CH2,
6-CH2/7-CH3; IR (cm−1) 3510(br s) (ν OH), 2980(s), 2870(s),
1725(s) (ν CO), 1645(w) (ν CC), 1455(s), 1375(s), 1275(s) (ν
C−O−C ester), 1215(s), 1145(s), 1105(s), 1070(s); Calcd. for
C22H37O6 ([M + H]+) 397.25847, found 397.25820; C22H36O6, M =
396.52 g/mol.
(3) (a) Trost, B. M. Science 1991, 254, 1471−1477. (b) Trost, B. M.
Angew. Chem., Int. Ed. Engl. 1995, 34, 259−281. (c) Trost, B. M. Acc.
Chem. Res. 2002, 35, 695−705.
(4) Nickel: (a) Muller, H.; Wittenberg, D.; Seibt, H.; Scharf, E.
̈
Angew. Chem. 1965, 77, 318−323. (b) Radetich, B.; RajanBabu, T. V.
J. Am. Chem. Soc. 1998, 120, 8007−8008. (c) Neas, D.; Tursky, M.;
́
́
Tilerova, I.; Kotora, M. New J. Chem. 2006, 30, 671−674. (d) Boing,
̈
C.; Hahne, J.; Francis, G.; Leitner, W. Adv. Synth. Catal. 2008, 350,
1073−1080.
(5) Ruthenium: (a) Yamamoto, Y.; Ohkoshi, N.; Kameda, M.; Itoh,
K. J. Org. Chem. 1999, 64, 2178−2179. (b) Cetinkaya, B.; Demir, S.;
̈
Ozdemir, I.; Toupet, L.; Semeril, D.; Bruneau, C.; Dixneuf, P. H. New
J. Chem. 2001, 25, 519−521. (c) Yamamoto, Y.; Nakagai, Y.-i.;
Ohkoshi, N.; Itoh, K. J. Am. Chem. Soc. 2001, 123, 6372−6380.
̈
(d) Cietinkaya, B.; Demir, S.; Ozdemir, I.; Toupet, L.; Semeril, D.;
Bruneau, C.; Dixneuf, P. H. Chem.Eur. J. 2003, 9, 2323−2330.
(e) Michaut, M.; Santelli, M.; Parrain, J.-L. Tetrahedron Lett. 2003, 44,
2157−2159. (f) Yamamoto, Y.; Nakagai, Y.-i.; Itoh, K. Chem.Eur. J.
2004, 10, 231−236. (g) Terada, Y.; Arisawa, M.; Nishida, A. J. Org.
Chem. 2006, 71, 1269−1272. (h) Arisawa, M.; Terada, Y.; Takahashi,
K.; Nakagawa, M.; Nishida, A. J. Org. Chem. 2006, 71, 4255−4261.
(i) Fairlamb, I. J. S.; McGlacken, G. P.; Weissberger, F. Chem.
Commun. 2006, 988−990. (j) Miao, X.; Feder-Kubis, J.; Fischmeister,
C.; Pernak, J.; Dixneuf, P. H. Tetrahedron 2008, 64, 3687−3690.
(6) Platinum: (a) Kerber, W. D.; Koh, J. H.; Gagne, M. R. Org. Lett.
2004, 6, 3013−3015. (b) Kerber, W. D.; Gagne, M. R. Org. Lett. 2005,
7, 3379−3381. (c) Feducia, J. A.; Campbell, A. N.; Doherty, M. Q.;
Gagne, M. R. J. Am. Chem. Soc. 2006, 128, 13290−13297. (d) Bell, F.;
Holland, J.; Green, J. C.; Gagne, M. R. Organometallics 2009, 28,
2038−2045.
N
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(7) Rhodium: (a) Bright, A.; Malone, J. F.; Nicholson, J. K.; Powell,
J.; Shaw, B. L. J. Chem. Soc., Chem. Commun. 1971, 712−713.
(b) Grigg, R.; Mitchell, T. R. B.; Ramasubbu, A. J. Chem. Soc., Chem.
Commun. 1980, 27−28. (c) Sato, Y.; Oonishi, Y.; Mori, M.
Organometallics 2003, 22, 30−32. (d) Oonishi, Y.; Taniuchi, A.;
Mori, M.; Sato, Y. Tetrahedron Lett. 2006, 47, 5617−5621. (e) Oonishi,
Y.; Saito, A.; Mori, M.; Sato, Y. Synthesis 2009, 969−979.
(23) Using [Pd(MeCN)4](BF4)2 at ambient temperature, screening
of various solvents (PhCH3, PhCF3, C6F6, CH3NO2, MeCN, THF,
Et2O) and an additive (CF3CH2OH) did not improve the chemo-
selectivity. We observed limited catalyst solubility in hydrocarbon
solvents and low conversion rates in Lewis basic solvents.
(24) Lloyd-Jones, G., C.; Stephen, S., C. Chem. Commun. 1998,
2321−2322.
(25) Sen, A.; Lai, T. W. Organometallics 1983, 2, 1059−1060.
(26) Attempts to isolate (crystallization) or characterize (NMR) Pd
species from the reaction mixture were unsuccessful.
(8) Titanium: (a) Okamoto, S.; Livinghouse, T. J. Am. Chem. Soc.
2000, 122, 1223−1224. (b) Okamoto, S.; Livinghouse, T. Organo-
metallics 2000, 19, 1449−1451.
(27) See Computational Chemistry.
(9) Palladium: (a) Schmitz, E.; Heuck, U.; Habisch, D. J. Prakt. Chem.
1976, 318, 471−478. (b) Schmitz, E.; Urban, R.; Heuck, U.;
(28) We are grateful to a reviewer for his insightful comments.
(29) The cycloisomerization of 1b using our standard protocol
([Pd(MeCN)4](BF4)2, CH2Cl2, ambient temperature) at different
concentrations (c = 0.5 M, 0.1 M, 0.01 M) did not alter the monomer/
dimer ratio.
Zimmermann, G.; Grundemann, E. J. Prakt. Chem. 1976, 318, 185−
̈
192. (c) Grigg, R.; Mitchell, T. R. B.; Ramasubbu, A. J. Chem. Soc.,
Chem. Commun. 1979, 669−670. (d) Grigg, R.; Malone, J. F.; Mitchell,
T. R. B.; Ramasubbu, A.; Scott, R. M. J. Chem. Soc., Perkin Trans. 1
1984, 1745−1754. (e) Takacs, J. M.; Zhu, J.; Chandramouli, S. J. Am.
Chem. Soc. 1992, 114, 773−774. (f) Heumann, A.; Moukhliss, M.
Synlett 1998, 1211−1212. (g) Widenhoefer, R. A.; Perch, N. S. Org.
Lett. 1999, 1, 1103−1105. (h) Kisanga, P.; Widenhoefer, R. A. J. Am.
Chem. Soc. 2000, 122, 10017−10026. (i) Bray, K. L.; Charmant, J. P.
H.; Fairlamb, I. J. S.; Lloyd-Jones, G. C. Chem.Eur. J. 2001, 7, 4205−
4215. (j) Bray, K. L.; Fairlamb, I. J. S.; Lloyd-Jones, G. C. Chem.
Commun. 2001, 187−188. (k) Bray, K. L.; Lloyd-Jones, G. C. Eur. J.
Org. Chem. 2001, 1635−1642. (l) Bray, K. L.; Fairlamb, I. J. S.; Kaiser,
J.-P.; Lloyd-Jones, G. C.; Slatford, P. A. Top Catal. 2002, 19, 49−59.
(m) Fairlamb, I. J. S.; Grant, S.; Tommasi, S.; Lynam, J. M.; Bandini,
M.; Dong, H.; Lin, Z.; Whitwood, A. C. Adv. Synth. Catal. 2006, 348,
2515−2530. (n) Song, Y.-J.; Jung, I. G.; Lee, H.; Lee, Y. T.; Chung, Y.
K.; Jang, H.-Y. Tetrahedron Lett. 2007, 48, 6142−6146. (o) Feng, L.;
Gan, Z.; Nie, X.; Sun, P.; Bao, J. Catal. Commun. 2010, 11, 555−559.
(10) Ziegler, K. Angew. Chem. 1956, 68, 721−729.
(30) The E-configuration of the C1/C2 double bond of 6b was
assigned on the basis of the 1H NMR coupling constant of the vinylic
protons (3J = 15.3 Hz). The cycloisomerization/dimerization of
( )-syn-1b could generate up to four racemic diastereomers. Notably,
1
only two major diastereomers were detected by H NMR spectros-
copy. In those cases, where the assignment of the relative configuration
of the dimers was successful, we observed a 2′,3′-cis-relationship.
(11) (a) Mach, K.; Sedmera, P.; Petrusova, L.; Antropiusova, H.;
Hanus, V.; Turecek, F. Tetrahedron Lett. 1982, 23, 1105−1108.
(b) Lehmkuhl, H.; Tsien, Y.-L. Chem. Ber. 1983, 116, 2437−2446.
(c) Akita, M.; Yasuda, H.; Nagasuna, K.; Nakamura, A. Bull. Chem. Soc.
Jpn. 1983, 56, 554−558. (d) Piers, W. E.; Shapiro, P. J.; Bunel, E. E.;
Bercaw, J. E. Synlett 1990, 74−84. (e) Bazan, G. C.; Rodriguez, G.;
Ashe, A. J.; Al-Ahmad, S.; Kampf, J. W. Organometallics 1997, 16,
2492−2494. (f) Thiele, S.; Erker, G. Chem. Ber. 1997, 130, 201−207.
(12) (a) Piers, W. E.; Bercaw, J. E. J. Am. Chem. Soc. 1990, 112,
9406−9407. (b) Molander, G. A.; Hoberg, J. O. J. Am. Chem. Soc.
1992, 114, 3123−3125. (c) Negishi, E.; Jensen, M. D.; Kondakov, D.
Y.; Wang, S. J. Am. Chem. Soc. 1994, 116, 8404−8405.
(d) Shaughnessy, K. H.; Waymouth, R. M. J. Am. Chem. Soc. 1995,
117, 5873−5874.
(31) The cycloisomerization/dimerization/cycloisomerization of
( )-syn-1b could generate up to eight racemic diastereomers. Notably,
only two diastereomers of ( )-7b were detected. The major
diastereomer could be isolated and fully characterized, and its all-cis
configuration was verified by 2D NOESY experiments; in accordance
with the assumed S2 (enantiotopic atoms or groups of atoms) or C2
symmetry (homotopic atoms or groups of atoms) of this diastereomer
of ( )-7b, the corresponding NMR spectra display a half set of
resonances.
(13) (a) Bogdanovic, B.; Henc, B.; Karmann, H.-G.; Nussel, H.-G.;
̈
Walter, D.; Wilke, G. Ind. Eng. Chem. 1970, 62, 34−44.
(b) Bogdanovic, B. Adv. Organomet. Chem. 1979, 17, 105−140.
(14) Behr, A.; Freudenberg, U.; Keim, W. J. Mol. Catal. 1986, 35,
9−17.
(32) See the Supporting Information for Ortep presentation; CCDC
868298 contains the supplementary crystallographic data for this
contribution. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
(33) (a) Beesley, R. M.; Ingold, C. K.; Thorpe, J. F. J. Chem. Soc.,
Trans. 1915, 107, 1080−1106. (b) Sammes, P. G.; Weller, D. J.
Synthesis 1995, 1205−1222. (c) Jung, M. E.; Piizzi, G. Chem. Rev.
2005, 105, 1735−1766.
(34) See the Supporting Information for Ortep presentation; CCDC
868297 contains the supplementary crystallographic data for this
contribution. These data can be obtained free of charge from The
Cambridge Crystallographic Data Centre via www.ccdc.cam.ac.uk/
data_request/cif.
(35) Frisch, M. J.; et al. Gaussian 03, Revision E.01; Gaussian, Inc.:
Wallingford CT, 2004. Full reference given in the Supporting
Information.
(15) Walther, D.; Dohler, T.; Heubach, K.; Klobes, O.; Schweder, B.;
̈
Gorls, H. Z. Anorg. Allg. Chem. 1999, 625, 923−932.
̈
(16) Kisanga, P.; Goj, L. A.; Widenhoefer, R. A. J. Org. Chem. 2001,
66, 635−637.
(17) Bray, K. L.; Lloyd-Jones, G. C.; Munoz, M. P.; Slatford, P. A.;
Tan, E. H. P.; Tyler-Mahon, A. R.; Worthington, P. A. Chem.Eur. J.
2006, 12, 8650−8663.
(18) Tan, E. H. P.; Lloyd-Jones, G. C.; Harvey, J. N.; Lennox, A. J. J.;
Mills., B. M. Angew. Chem., Int. Ed. Engl. 2009, 48, 6262−6265.
(19) Tan, E. H. P.; Lloyd-Jones, G. C.; Harvey, J. N.; Lennox, A. J. J.;
Mills, B. M. Angew. Chem., Int. Ed. Engl. 2011, 50, 9602−9606.
(20) Nelson, B.; Hiller, W.; Pollex, A.; Hiersemann, M. Org. Lett.
2011, 13, 4438−4441.
(21) Schramm, R. F.; Wayland, B. B. Chem. Commun. 1968, 898−
899.
(22) A possible explanation of this experimental observation is that
the generation of the active catalyst, likely a “LnPd−H”-species, from
the precatalyst requires residual quantities of water.
(36) (a) Becke, A. D. J. Chem. Phys. 1993, 98, 5648−5652. (b) Lee,
C. T.; Yang, W.; Parr, R. G. Phys. Rev. B 1988, 37, 785−789.
O
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX

The Journal of Organic Chemistry
Article
(c) Miehlich, B.; Savin, A.; Stoll, H.; Preuss, H. Chem. Phys. Lett. 1989,
157, 200−206.
(37) Friesner, R. A.; Murphy, R. B.; Beachy, M. D.; Ringnalda, M. N.;
Pollard, W. T.; Dunietz, B. D.; Cao, Y. J. Phys. Chem. A 1999, 103,
1913−1928.
(38) (a) Weigend, F.; Ahlrichs, R. Phys. Chem. 2005, 7, 3297−3305.
(b) Andrae, D.; Haußermann, U.; Dolg, M.; Stoll, H.; Preuß, H. Theor.
̈
Chim. Acta 1990, 77, 123−141.
(39) Peterson, K. A.; Figgen, D.; Goll, E.; Stoll, H.; Dolg, M. J. Chem.
Phys. 2003, 119, 11113−11123.
(40) In context of diene-cycloisomerization/polymerization, a
number of similar PdII-O-donor-chelate complexes, acting as resting
states for these transformations, have been described in the literature:
(a) Goj, L. A.; Widenhoefer, R. A. J. Am. Chem. Soc. 2001, 123,
11133−11147. (b) Goj, L. A.; Cisneros, G. A.; Yang, W.; Widenhoefer,
R. A. J. Organomet. Chem. 2003, 687, 498−507. (c) DiRenzo, G. M.;
White, P. S.; Brookhart, M. J. Am. Chem. Soc. 1996, 118, 6225−6234.
(d) Margl, P.; Ziegler, T. J. Am. Chem. Soc. 1996, 118, 7337−7344.
(e) Mecking, S.; Johnson, L. K.; Wang, L.; Brookhart, M. J. Am. Chem.
Soc. 1998, 120, 888−899. (f) Perch, N. S.; Widenhoefer, R. A.
Organometallics 2001, 20, 5251−5253. (g) Nakamura, A.; Munakata,
K.; Ito, S.; Kochi, T.; Chung, L. W.; Morokuma, K.; Nozaki, K. J. Am.
Chem. Soc. 2011, 133, 6761−6779. (h) Perch, N. S.; Widenhoefer, R.
A. J. Am. Chem. Soc. 2004, 126, 6332−6346.
(41) We were unable to locate a static stationary point between
TS(M11−M13) and M13.
(42) (a) Hiersemann, M. Tetrahedron 1999, 55, 2625−2638.
(b) Hiersemann, M.; Lauterbach, C.; Pollex, A. Eur. J. Org. Chem.
1999, 2713−2724. (c) Hiersemann, M. Synlett 2000, 415−417.
(d) Hiersemann, M. Eur. J. Org. Chem. 2001, 483−491. (e) Hiersemann,
M.; Abraham, L.; Pollex, A. Synlett 2003, 1088−1095.
(43) Juaristi, E.; Martinez-Richa, A.; Garcia-Rivera, A.; Cruz-Sanchez,
J. S. J. Org. Chem. 1983, 48, 2603−2606.
(44) Kirsten, M.; Rehbein, J.; Hiersemann, M.; Strassner, T. J. Org.
Chem. 2007, 72, 4001−4011.
(45) Rehbein, J.; Leick, S.; Hiersemann, M. J. Org. Chem. 2009, 74,
1531−1540.
(46) Hiersemann, M. Synthesis 2000, 1279−1290.
(47) Sen, A.; Lai, T. W. Organometallics 1983, 2, 1059−1060.
(48) (a) Schramm, R. F.; Wayland, B. B. Chem. Commun. 1968, 898−
899. (b) Lai, T. W.; Sen, A. Organometallics 1984, 3, 866−870.
P
dx.doi.org/10.1021/jo3004088 | J. Org. Chem. XXXX, XXX, XXX−XXX