Journal Name
ARTICLE

This document is the accepted manuscript version of a published work that appeared in final form
in Cat. Sci. Technol. 2018, 5251-5258. DOI: 10.1039/c8cy00684a.
Received 00th January 20xx,
Accepted 00th January 20xx
DOI: 10.1039/x0xx00000x

Ni-Catalysed Intramolecular [4+4]-Cycloadditions of Bis-dienes
towards Eight-membered Fused Bicyclic Systems: A Combined
Experimental and Computational Study
Nuria Llorente,aǂ Héctor Fernández-Pérez,aǂ Antonio Bauzá,c Antonio Frontera*c and Anton VidalFerran*a,b
Detailed investigations on the use of nickel(0)-based catalysts for intramolecular [4+4]-cycloadditions are presented.
Nickel(0) complexes derived from electron-rich triarylphosphines proved to be efficient catalysts for intramolecular [4+4]cycloadditions of an array of structurally diverse bis-dienes (10 examples, up to 78% isolated yield). The reported synthetic
methodology leads to cis-eight-membered fused [6.3.0] bicyclic compounds as well as trans- or cis- eight-membered fused
[6.4.0] bicyclic systems. Computational studies on the stereo-determining step of the reaction in combination with
experimental results demonstrated that the stereochemical outcome is dictated by the length of the chain linking the two
diene units and the geometry of the C=C double bonds of the substrates.

Introduction
Medium-sized cyclic compounds, and in particular eightmembered fused polycyclic systems, are key structural motifs in
a significant number of natural products. 1 Among the different
strategies for the preparation of such compounds, 2 metalmediated intramolecular [4+4]-cycloadditions are an elegant,
efficient, and atom-economical synthetic methodology.
Although metal-mediated [4+4]-cycloadditions of two diene
units have been known for some time, 3 it was Wender’s
research group4 who pioneered the use of intramolecular
nickel-catalysed
[4+4]-cycloadditions
for
the
highly
diastereoselective synthesis of complex natural products (or
advanced synthetic intermediates thereof). In Wender’s
studies, the configurations of the substituted carbons in the
tether linking the two diene units were responsible for the high
diastereoselectivities observed.4 The practicality of this
approach was demonstrated by preparing (+)-asteriscanolide
and (±)-salsolene oxide using this chemistry as the key step. 5
More recently, Cheung and co-workers have used rhodium(I)

complexes as catalysts in this type of cycloaddition for the
efficient preparation of the corresponding cyclooctadiene
derivatives,6 though harsher reaction conditions than those
reported by Wender are required.
Whilst diastereoselective nickel-catalysed intramolecular
[4+4]-cycloadditions of bis-dienes have been reported, the
study of this nickel chemistry on prochiral bis-dienes containing
heteroatoms in the tether between the two diene units
remained scarcely explored.4 Herein, we wish to report our
findings on the development of nickel(0)-based achiral catalysts
for intramolecular [4+4]-cycloadditions of structurally diverse
prochiral bis-diene substrates (Scheme 1). We also aim to
disclose the effects of substituents and C=C double bond
geometries on the outcome of the [4+4]-cycloadditions.
Computational studies have been performed to identify the
relevant transition states of the stereo-determining step and
their relative stabilities and consequently provide a
rationalisation of the stereochemical outcome of the reaction.

Results and Discussion
a. Institute

of Chemical Research of Catalonia (ICIQ) & The Barcelona Institute of
Science and Technology (BIST), Av. Països Catalans 16, 43007 Tarragona, Spain. Email: avidal@iciq.cat
b. ICREA, Passeig Lluís Companys 23, 08010 Barcelona, Spain.
c. Departament de Química, Universitat de les Illes Balears (UIB), Ctra. de
Valldemossa km 7.5, 07122 Palma de Mallorca, Spain.
† Electronic Supplementary Information (ESI) available: Experimental procedures,
spectral data for known and new compounds, selected crystallographic information
and cartesian coordinates of the calculated structures. CCDC 1834976 and 1834977
contains the supplementary crystallographic data for this paper. See DOI:
10.1039/x0xx00000x.
ǂ These authors contributed equally to this work.

Study of the [4+4]-cycloadditions
At the onset of our studies, we used oxygen-containing bisdiene 1a as the model substrate, as its preparation was already
reported.7 The intramolecular [4+4]-cycloaddition of 1a was

J. Name., 2018, 00, 1-3 | 1

ARTICLE

Journal Name
Table 1 Screening of ligands for the nickel-catalysed intramolecular [4+4]-cycloaddition
of substrate 1aa

Scheme 1 Nickel-catalysed intramolecular [4+4]-cycloadditions studied

studied in the presence of catalytic amounts of [Ni(cod)2] (cod =
1,5-cyclooctadiene) and a set of phosphorus monodentate
ligands (L1L6; see Table 1 for the structures), in order to test
the influence of the ligand on both the reactivity and selectivity
of the nickel catalyst.8
When PPh3 (L1) was used as the ligand, the cis-fused
cyclooctadiene 2a was obtained in good isolated yields (see
entry 1 in Table 1). Under these reaction conditions, one of the
possible [4+2]-cycloaddition products9 was identified in the
crude mixture. We were then interested in studying how the
electronic properties of the phosphorus ligands affected the
outcome of the cycloaddition reactions. Interestingly, the
electron-rich triarylphosphine L2 provided compound 2a in
almost identical yield to triphenylphosphine and gave only
minimal amounts of products 3a (compare entry 2 with entry 1
in Table 1). On the contrary, the conversion of the reaction and
selectivity towards product 2a dropped when the reaction was
conducted with ligand L3, which incorporates sigma electron
withdrawing groups (compare entry 3 with entries 12 in Table
1). Likewise, the use of monophosphite ligand L4 did not
significantly improve the previous results and led to decreased
activity and yield for cyclooctadiene 2a (see entry 4 in Table 1).
As far as the phosphoramidite ligand L5 is concerned, the
corresponding nickel complex proved to be an

Entry

Ligand

1
2
3
4
5
6

L1
L2
L3
L4
L5
L6

Conv.
(%)b
>99
98
97
74
>99
38

Yield
cis-2a (%)b
58 (52)
57 (56)
28
42
56 (46)
10

Yield
3a (%)b
6
<1
8
22 (16)
-

Yield
4a (%)b
2
<1
-

a

The results are the average of at least two independent runs. b Calculated by 1H
NMR spectroscopy of the crude mixture using 1,3,5-trimethoxybenzene as internal
standard. In parentheses: isolated yields after column chromatography on silica gel
impregnated with AgNO3 (10%).

active catalyst for the intramolecular [4+4]-cycloaddition of
substrate 1a, with full conversion of 1a obtained and the
product 2a isolated in a moderate yield (46%, see entry 5 in
Table 1). It is interesting to note that L5 allowed notable
amounts of one of the possible [4+2]-cycloadducts 3a to be
isolated and characterised.10 Finally, trialkylphosphine ligand L6
was assessed in this initial screening. As shown in entry 6 (Table
1), low reactivity was observed towards the product 2a.
The previous results demonstrate that changes in the
electronic nature of the ligands lead to significant differences in
the yield and selectivity of the cycloaddition reactions.
Interestingly, electron-rich triphenylphosphine L2 was the most
selective ligand for the nickel-catalysed intramolecular [4+4]cycloaddition of bis-diene 1a and was selected for further
catalytic studies. Thus, with catalytic conditions in hand, we
turned our attention to assessing the scope of the nickelcatalysed intramolecular [4+4]-cycloaddition over a set of
structurally diverse prochiral bis-diene substrates. First, we
focused our attention on bis-dienes 1b1f,7 substrates that lead
to 10-substituted bicyclo[6.3.0]-undeca-2,6-diene systems (see
Table 2 for the structures). The results summarised in Table 2
indicate that the nickel-catalysed cycloadditions to eightmembered fused bicyclic systems were successfully
accomplished in almost all cases. Replacement of the oxa group
in 1a by a 1,3-dimethoxy-1,3-dioxopropan-2-yl or aza

2 | J. Name., 2018, 00, 1-3

Journal Name
substituent was well tolerated and the corresponding [4+4]cycloadducts were isolated in satisfactory yields (see entries 1
and 2 in Table 2). Analysis of the NMR spectra of the eightmembered fused bicyclic systems derived from 1a1c revealed
a cis-fusion between the five- and eight-membered rings. This
type of ring fusion was in agreement with previous results in the
literature,4a and was confirmed by X-ray analysis of a single
crystal of compound 2c.11
We also investigated the outcome of the reaction upon
placing a methyl group at the 1-, 2- or 3-position of one of the
diene units. Thus, intramolecular [4+4]-cycloadditions of
substrates 1d1f were examined. As shown in Table 2, the
selectivities of the cycloadditions were heavily influenced by the
substitution pattern of the diene. Introduction of a methyl
group at the terminal position of the diene (substrate 1d) was
detrimental for the yield: compound 2d was the only adduct
that could be isolated in diastereomerically pure form, but in
low isolated yield (13%, see entry 3 in Table 2). The relative
configuration of the three stereogenic carbons in 2d was
established by extensive 1H-selective decoupling, 1D NOE and
2D NMR experiments.11 Substitution of the diene at the 2position (substrate 1e) was detrimental, as the desired [4+4]cycloadduct could not be detected in the complex mixture of
products that was obtained (see entry 4 in Table 2). Finally,
substrate 1f bearing a methyl group at the 3-position cyclised to
the cis-fused bicyclic product 2f in moderate yield (up to 35%
isolated yield, see entry 5 in Table 2). The relative configuration
of the two stereogenic carbons in 2f was established by
extensive NMR experiments.11

ARTICLE
Table 2 Nickel-catalysed intramolecular [4+4]-cycloaddition towards bicyclo[6.3.0] ring
systemsa

a

The results are the average of at least two independent runs. Reactions
performed in the presence of 10.0 mol% [Ni(cod) 2] and 21.0 mol% L2 in
toluene at 60 ºC for 24h. c 31 mol% L2 used as ligand. d Not determined. b See
note b in Table 1.

Encouraged by these results, we extended the use of this
[4+4]-cycloaddition strategy to the preparation of 10substituted bicyclo[6.4.0]dodeca-2,6-diene systems starting
from the corresponding bis-dienes 1g1i (see Table 3 for the
structures). The required substrates were synthesised following
well-established synthetic protocols,7 though it should be noted
that substrate 1i could only be prepared as a mixture of isomers
(E,E-1i:E,Z-1i ratio of 88:12), which were inseparable by
standard column chromatography. Under our optimised
reaction conditions, substrate 1g gave [4+4]-cycloadduct 2g (up
to 67% isolated yield, see entry 1 in Table 3). Spectroscopic data
of the cycloaddition product pointed to the trans-eightmembered fused bicyclic system 2g. This type of ring fusion was
in agreement with previous results in the literature for
comparable compounds,4c and was unambiguously confirmed
by X-ray analysis (Fig. 1).
The same trend was observed for the intramolecular [4+4]cycloaddition of substrates 1h and 1i, which contained oxa and
aza-substituted groups in the chain linking the two diene units.
The eight-membered fused bicyclic systems were obtained in
good isolated yields (61% and 67% for 1h and 1i, respectively;
see entries 2 and 3 in Table 3). The [4+4]-cycloadducts derived
from 1i were obtained as an inseparable mixture of two
isomeric compounds in an 88:12 ratio. It is interesting to note
the close relationship between the E,E:E,Z isomer ratio of the
starting material 1i and the ratio of the isomeric [4+4]-

J. Name., 2018, 00, 1-3 | 3

ARTICLE

Journal Name

Table 3 Nickel-catalysed intramolecular [4+4]-cycloaddition towards bicyclo[6.4.0] ring
systemsa

Scheme 2 Stereospecific nickel-catalysed intramolecular [4+4]-cycloaddition of 1i

Computational studies

a, b

See notes a and b in Table 1

cycloadducts (trans-2i and cis-2i), thus suggesting a direct
relationship between the geometry of the diene units and the
structure of the final products.
In order to fully understand the outcome of the
cycloaddition reaction of 1i, the separation of E,E-1i and E,Z-1i
was attempted. The two isomers of 1i could not be separated
by standard column chromatography, but pure samples of E,E1i and E,Z-1i could be obtained by semipreparative HPLC. 11 With
E,E-1i and E,Z-1i in hand, we examined the Ni-catalysed
intramolecular [4+4]-cycloadditions employing [Ni(cod)2] and
L2 as catalyst. As shown in Scheme 2, each isomer of substrate
1i exclusively led to one [4+4]-cycloadduct. Whilst E,E-1i
evolved under the reaction conditions to the trans-eightmembered fused bicyclic system 2i in high isolated yield (78%),
E,Z-1i led to the cis-fused analogue (55%). Structural
assignments were made by analogy with the spectroscopic data
of trans-fused product 2g and by performing extensive NMR
experiments on cis-2i. These results illustrate the importance of
the geometry of the diene on the outcome of the cycloaddition,
as E,E-1i and E,Z-1i lead to different diastereoisomeric
cycloadducts (i.e., trans-2i or cis-2i, respectively).

Fig. 1 X-ray crystal structure of trans-2g (ORTEP drawings showing thermal
ellipsoids at 50% probability).

4 | J. Name., 2018, 00, 1-3

To shed light on the outcome of the [4+4]-cycloaddition
reactions, we performed a theoretical investigation (BP-86D3/def2-TZVP, see ESI for details) into the reactivity of
substrates 1a, E,E-1h and E,Z-1h. A tentative reaction pathway
for the intramolecular nickel-catalysed [4+4]-cycloadditions of
bis-dienes is provided in Scheme 3. The accepted view for this
transformation3b,3c,3f,4a,4c,12 is that the nickel precursor reacts
with the substrate to form bis-diene nickel complexes 5 that,
after an oxidative cyclisation of the two internal C=C bonds, lead
to bis-allyl Ni(II) complexes 6. Reductive elimination from these
Ni(II)-complexes 6 leads to the cyclised product and nickel
complexes capable of re-entering the catalytic cycle. We have
focused our study on the first C–C bond formation, where the
five- (for substrate 1a) or six-membered ring (for compounds
E,E-1h and E,Z-1h) is formed and the stereochemistry of the
ring-fusion is defined. We have compared the energies of the
transition states (TSs) for the paths leading to the cis- and transcycloadducts for the three substrates incorporated to this
computational study.

Journal Name
Scheme 3 Tentative reaction pathway
intramolecular [4+4]-cycloadditions

ARTICLE
rationale

for

nickel-catalysed

We first studied the reactivity of compound 1a (see Fig. 2).
Experimentally, phosphine L2 is used in the reaction. However,
our computational studies did not incorporate L2 in the
coordination sphere of the Ni(0) centre as by coordinating to
the four double bonds of the substrate it had already
accommodated eighteen electrons. We performed a

conformational search on the starting complex [Ni(1a)] and
found two possible binding modes (denoted as pre-TS-1 and
pre-TS-2) depending on the relative orientation of the diene
units. Remarkably, pre-TS-2 is considerably more stable than
pre-TS-1 and it has the adequate orientation to yield the transisomer. Both transition states on the potential hypersurface

Fig. 2 Optimised geometries BP86-D3/def2-TZVP level of theory of the pre-transition states (pre-TS), transition states (TS), intermediates (I) and products for the reaction of 1a and
its energetic profile (kcal/mol) in toluene (distances in Å).

Fig. 3 Optimised geometries BP86-D3/def2-TZVP level of theory of the pre-transition states (pre-TS), transition states (TS), intermediates (I) and products for the reaction of E,E-1h
and its energetic profile (kcal/mol) in toluene (distances in Å).

were located and the TS that yields the cis-isomer (denoted as
TS-1) is lower in energy than the TS for the trans-isomer (TS-2;
see Fig. 2), which is in good agreement with the experimental
results.13 The reaction barrier is 18.8 kcal/mol for TS-1 and 26.3
kcal/mol for TS-2. The higher stability of TS-1 (i.e. 7.5 kcal/mol)
can be attributed to the closer proximity of the metal centre to

the C atoms that are forming the C–C bond in TS-1 (dNi···C = 2.21
Å) than in TS-2 (dNi···C = 2.85 Å). Consequently, the more stable
pre-TS-2 trans-isomer needs to isomerize to the less stable preTS-1 cis-isomer prior to the C–C bond formation. In the resulting
intermediates (denoted as I-1 and I-2 in Fig. 2) the nickel(II)
centre is able to incorporate the phosphine as an additional

J. Name., 2018, 00, 1-3 | 5

ARTICLE

Journal Name

ligand, stabilising the complexes. The reaction is highly
exergonic in both cases.

Fig. 4 Optimised geometries BP86-D3/def2-TZVP level of theory of the pre-transition states (pre-TS), transition states (TS), intermediates (I) and products for the reaction of E,Z-1h
and its energetic profile (kcal/mol) in toluene (distances in Å).

We have also studied the reactivity of compound E,E-1h (see
Fig. 3). Experimentally, the cycloaddition of this substrate leads
to trans-2h rather than cis-2h (it is interesting to note that the
substrate leading to a cis-fusion between the five- and eightmembered rings, i.e. E,E-1a, and that leading to a trans-fusion
between the six- and eight-membered rings, i.e. E,E-1h, differ in
the length of the spacer used to connect both π-systems).
Similarly to compound E,E-1a, we have also performed a
conformational study of [Ni(E,E-1h)] and the trans-isomer is
also the most stable (5.8 kcal/mol, denoted as pre-TS-3 in Fig.
3). We located both transition states on the potential
hypersurface, and, as for the cycloadditions of 1a, we
considered that the phosphine does not participate in the
process (transition states for the cycloaddition of E,E-1h with
one uncoordinated terminal C=C bond of the substrate and
ligand L2 coordinated to the Ni(0) centre were located and were
higher in energy than those involving coordination of the
substrate through the four C=C bonds13). Gratifyingly, the TS
that yields the trans isomer (TS-3) is lower in energy than the TS
for the cis isomer (TS-4; see Fig. 3), which is in good agreement
with the experimental results. The reaction barrier is 22.3
kcal/mol for TS-3 and 27.7 kcal/mol for TS-4. In this case TS-3 is
more stable than TS-4 because the former can adopt a perfect
chair conformation in the formation of the six-membered ring
(see Figure S3 for a detailed representation of the TSs). In
contrast, the TS-4 adopts a boat-like conformation in the C–C
bond formation. The reaction is highly exergonic for both
stereoisomers upon coordination of the phosphine.

6 | J. Name., 2018, 00, 1-3

In order to support the level of theory used herein, we have
also computed the energies of the TSs for compounds 1a and
E,E-1h using MP2/def2-TZVP (in toluene) and we have observed
the same trend. That is, the TS-1 is lower than the TS-2 for 1a in
5.6 kcal/mol (7.5 kcal/mol using DFT) and the TS-3 is lower than
the TS-4 for E,E-1h in 2.3 kcal/mol (5.4 kcal/mol using DFT).14
Finally, we studied the most favourable [4+4]-cycloaddition
reaction pathway for E,Z-1h, which has the same spacer as E,E1h but has one (Z)-configured C=C double bond instead of two
(E)-configured C=C bonds. Although the cycloaddition reaction
of this substrate was not studied experimentally, 15 we carried
out computational studies with E,Z-1h rather than E,Z-1i for the
sake of consistency with the other bis-dienes studied at the
computational level. In this case, the conformational study of
[Ni(E,Z-1h)] revealed that both isomers (pre-TS-5 and pre-TS-6)
were almost isoenergetic. Moreover, we could not locate any
transition state with the four C=C bonds coordinated to the
nickel centre: the Z-configuration of the double bond in E,Z-1h
hinders the coordination of its terminal double bond to the
nickel centre and facilitates the coordination of L2. Thus, we
have located the corresponding transition states with L2 and
three C=C bonds coordinated to the nickel centre on the
potential hypersurface and, remarkably, the TS that leads to the
cis isomer (TS-5) is slightly lower in energy than that for the
trans-isomer (TS-6; see Fig. 4). The reaction barrier is 20.6
kcal/mol for TS-5 and 21.8 kcal/mol for TS-6. It should be noted
that whilst the computed results are in qualitative agreement
with the experimental results, they do not quite match the

Journal Name
relative magnitudes of the stereoselection (only one
stereoisomer is experimentally observed from E,Z-1h, whilst the
energy differences between TS-5 and TS-6 is ca. 1 kcal/mol). In
spite of TS-6 adopting a perfect chair conformation for the
formation of the six-membered ring (see Fig. S4 for a detailed
representation of the transition states), TS-5 presents an extra
stabilisation energy that comes from the interaction of a Hatom of the uncoordinated double bond with a C=C bond of the
aromatic ring (see dashed lines in Fig. 4). The non-coordinated
double bond in TS-6 is oriented totally different and does not
establish any interaction with the phosphine. It should be noted
that the reaction is exergonic for the cis isomer (–4.1 kcal/mol)
and thermo-neutral for the trans isomer (–0.1 kcal/mol) (see
Fig. 4). These data therefore suggest that the ring fusion is
thermodynamically rather than kinetically controlled.
Consequently, the formation of E,Z-6h-π,π-trans is reversible
and the selectivity can be explained by the energy difference
between cis and trans isomers (–4.0 kcal/mol), which correlates
with the selectivity observed experimentally.

Conclusions
In
summary,
nickel-catalysed
intramolecular
[4+4]cycloadditions for the preparation of cis-eight-membered fused
[6.3.0] bicyclic compounds and trans- or cis- eight-membered
fused [6.4.0] bicyclic systems from structurally diverse bisdienes are reported. The catalytic intramolecular [4+4]cycloaddition reaction proceeds efficiently on a set of bis-dienes
linked by a three-atom chain to afford cis-configured eightmembered carbocycles fused to a five-membered ring.
Analogously, the reported chemistry for bis-dienes linked by a
four-atom chain gives access to trans- or cis-configured eightmembered carbocycles fused to a six-membered ring.
Computational studies on the stereo-determining step of the
reaction have demonstrated that the stereochemical outcome
of the reaction is dictated by the length of the chain linking the
two diene units, and the geometry of the C=C double bonds in
the diene units of the substrates. Ongoing research activities
are aimed at developing catalytic enantioselective versions of
this transformation.

Experimental
For the general considerations of the experimental and
theoretical methods, see the Electronic Supplementary
Information (ESI).
General procedure for [4+4]-cycloadditions
A solution of the substrate (1.00 mmol) and ligand (0.21 mmol)
in anhydrous and deoxygenated toluene was prepared under
inert atmosphere. [Ni(cod)2] (0.10 mmol) was added dropwise
from a stock solution in anhydrous toluene. In all cases, the
molar concentration of a given substrate in the reaction
medium was adjusted to 0.20 M. The resulting mixture was
carefully heated at 60 °C under an argon atmosphere and stirred
for 24 h. The reaction mixture was then allowed to reach room

ARTICLE
temperature and was exposed to air for 1 h. The mixture was
filtered through a short pad of silica and further eluted with
diethyl ether. The filtrate was concentrated in vacuo and the
resulting crude mixture was analysed by NMR spectroscopy.
Purification of the desired product was achieved by column
chromatography on silica gel impregnated with silver nitrate
(10%).
Product trans-2i
Product trans-2i was prepared following the general procedure
starting from substrate E,E-1i (0.107 g, 0.34 mmol), ligand L2
(25.5 mg, 0.071 mmol), and [Ni(cod)2] (9.5 mg, 0.034 mmol). It
was obtained as colourless liquid (0.084 g, 78% yield). 1H NMR
(500 MHz, CDCl3) δ 7.66 – 7.64 (m, 2H, Harom), 7.34 – 7.32 (m,
2H, Harom), 5.60 – 5.55 (m, 1H), 5.53 – 5.48 (m, 1H), 5.33 – 5.29
(m, 1H), 5.25 – 5.22 (m, 1H), 3.88 – 3.82 (m, 2H), 2.72 – 2.65 (m,
1H), 2.47 – 2.37 (m, 5H), 2.27 – 2.17 (m, 4H), 1.91 (dd, 2JH-H = 3JHH = 11.3 Hz, 1H), 1.85 – 1.81 (m, 1H), 1.51 – 1.42 (m, 1H) ppm.
13C{1H} NMR (125 MHz, CDCl ) δ 143.6 (C
3
q arom), 133.5 (Cq arom),
132.4 (CH=), 129.8 (CHarom), 129.5 (CH=), 129.4 (CH=), 127.9
(CH=), 127.8 (CHarom), 51.7 (CH2), 46.7 (CH2), 42.3 (CH), 41.8
(CH), 32.3 (CH2), 27.9 (CH2), 27.8 (CH2), 21.7 (CH3) ppm. HRMS
(ESI+): m/z calcd for C18H24NO2S [M+H]+ 318.1522, found
318.1519.
Product cis-2i
Product cis-2i was prepared following the general procedure
starting from substrate E,Z-1i (15.0 mg, 0.0473 mmol), ligand L2
(3.57 mg, 0.0099 mmol), and [Ni(cod)2] (1.33 mg, 0.0047 mmol).
It was obtained as colourless liquid (8.3 mg, 55% yield). 1H NMR
(400 MHz, CDCl3) δ 7.65 – 7.62 (m, 2H), 7.33 – 7.31 (m, 2H), 5.61
– 5.51 (m, 3H), 5.20 – 5.16 (m, 1H), 3.52 – 3.47 (m, 1H), 3.31 –
3.25 (m, 2H), 2.74 – 2.52 (m, 5H), 2.43 (s, 3H), 2.14 – 1.97 (m,
2H), 1.67 – 1.58 (m, 2H) ppm. 13C{1H} NMR (100 MHz, CDCl3) δ
143.5 (Cq arom), 133.4 (Cq arom), 132.1 (CH=), 129.8 (CHarom), 128.9
(CH=), 128.3 (CH=), 128.2 (CH=), 127.8 (CHarom), 49.7 (CH2), 45.6
(CH2), 39.6 (CH), 37.8 (CH), 29.1 (CH2), 29.0 (CH2), 26.9 (CH2),
21.7 (CH3) ppm. All spectroscopic data were in agreement with
those previously reported in the literature.6

Conflicts of interest
There are no conflicts to declare.

Acknowledgements
We thank MINECO (CTQ2014-60256-P, CTQ2017-89814-P,
CTQ2014-57393-C2-1-P and CTQ2017-85821-R FEDER funds
and Severo Ochoa Excellence Accreditation SEV-2013-0319) and
the ICIQ Foundation for financial support. N. Llorente thanks the
ICIQ Foundation for a predoctoral fellowship. We are grateful to
J. Benet-Buchholz for X-ray crystallographic data and Ms. R.
Somerville for proof reading the manuscript.

Notes and references

J. Name., 2018, 00, 1-3 | 7

ARTICLE

1 See, for example: (a) N. Blanchard and J. Eustache, in Metathesis
in Natural Product Synthesis: Strategies, Substrates and Catalysts,
eds. J. Cossy, S. Arseniyadis and C. Meyer, Wiley-VCH Verlag GmbH &
Co. KGaA, 2010, ch. 1, pp. 1-43; (b) Y. Wang and Z.-X. Yu, Acc. Chem.
Res., 2015, 48, 2288.
2 For general reviews, see: (a) G. Kaupp, Angew. Chem., Int. Ed.
Engl., 1992, 31, 422; (b) S. M. Sieburth and N. T. Cunard, Tetrahedron,
1996, 52, 6251; (c) Z.-X. Yu, Y. Wang and Y. Wang, Chem. - Asian J.,
2010, 5, 1072.
3 (a) H. W. B. Reed, J. Chem. Soc. 1954, 1931; (b) W. Brenner, P.
Heimbach, H. Hey, E. W. Müller and G. Wilke, Liebigs Ann. Chem.,
1969, 727, 161; (c) P. W. N. M. Van Leeuwen and C. F. Roobeek,
Tetrahedron, 1981, 37, 1973; (d) H. tom Dieck and J. Dietrich, Chem.
Ber., 1984, 117, 694; (e) H. tom Dieck, J. Dietrich and G. Wilke,
Angew. Chem., Int. Ed. Engl., 1985, 24, 781; (f) A. Tenaglia, P. Brun
and B. Wägell, J. Organomet. Chem., 1985, 285, 343; (g) M. Mallien,
E. T. K. Haupt and H. tom Dieck, Angew. Chem., Int. Ed. Engl. 1988,
27, 1062; (h) K. U. Baldenius, H. Tom Dieck, W. A. König, D. Icheln and
T. Runge, Angew. Chem., Int. Ed. Engl., 1992, 31, 305.
4 (a) P. A. Wender and N. C. Ihle, J. Am. Chem. Soc., 1986, 108,
4678; (b) P. A. Wender and M. L. Snapper, Tetrahedron Lett., 1987,
28, 2221; (c) P. A. Wender and N. C. Ihle, Tetrahedron Lett., 1987, 28,
2451; (d) P. A. Wender and M. J. Tebbe, Synthesis, 1991, 1089; (e) P.
A. Wender, J. Nuss, D. B. Smith, A. Suarez-Sobrino, J. Vågberg, D.
Decosta and J. Bordner, J. Org. Chem., 1997, 62, 4908.
5 (a) P. A. Wender, N. C. Ihle and C. R. D. Correia, J. Am. Chem. Soc.,
1988, 110, 5904; (b) P. A. Wender, M. P. Croatt and B. Witulski,
Tetrahedron, 2006, 62, 7505.
6 J. W. Park, J. E. Park, J. H. Park, M. R. Hong, S. M. Kim, Y. K. Chung
and C. H. Kim, Synlett, 2016, 27, 455.
7 Its reported preparation method involved using (E)-5chloropenta-1,3-diene with small amounts of the cis-isomer (ratio
E,E:E,Z isomers > 95:5). For the preparation of substrate 1a, see also:
(a) R. Hertel, J. Mattay and J. Runsink, J. Am. Chem. Soc., 1991, 113,
657; for the preparation of substrate 1b, see also: (b) J. M. Takacs
and E. C. Lawson, Organometallics 1994, 13, 4787; for the
preparation of substrate 1c, see also: (c) M. Takimoto and M. Mori,
J. Am. Chem. Soc. 2002, 124, 10008.
8 The reaction of substrate 1a exclusively in the presence of
Ni(cod)2 (10.0 mol%) led to a complex mixture of unidentified
products at 90% conversion.

8 | J. Name., 2018, 00, 1-3

Journal Name

9 Given the small amounts of product, no attempts were made to
elucidate the relative stereochemistry of the stereogenic carbons.
10 In particular, the [4+2] cycloaddition product 3a using the
phosphoramidite ligand L5 corresponds to (3aR,4R,7aS)-4-vinyl1,3,3a,4,5,7a-hexahydroisobenzofuran (see Figures S33 and S34 for
details), whose spectroscopic data are in good agreement with those
previously reported in the literature (see reference 7a).

11 See the Electronic Supplementary Information (ESI) for details.
12 K. Masanari and Y. Tamaru, in Modern Organonickel Chemistry,
ed. Y. Tamaru, Wiley-VCH Verlag GmbH & Co. KGaA: Weinheim,
2005, ch. 5, pp 137-170.
13 Transition states for the cycloaddition reaction of E,E-1h with
with one uncoordinated terminal C=C bond of the substrate and
ligand L2 coordinated to the Ni(0) centre were located (see the
structures below and Figure S5 for details). The energies of the new
TSs (TS-5’cis and TS-6’trans, which lead to cis-2h and trans-2h,
respectively) were higher in energy (i.e., TS-5’cis = 27.3 kcal/mol and
TS-6’trans = 27.9 kcal/mol) than those involving coordination of the
substrate through the four C=C bonds) (i.e., TS-5cis = 22.3 kcal/mol
and TS-6trans = 27.7 kcal/mol).

14 The calculations at the MP2 level have not been performed in
E,Z-1i due to the size of the system incorporating L2 in the
calculations.
15 Substrate E,Z-1h is contained as a minor impurity in the samples
of substrate E,E-1h (< 5%, see reference 7). Accordingly, the [4+4]cycloadduct derived from E,Z-1h (i.e. cis-2h) was present in the
corresponding amounts in the cycloaddition reaction mixture and
could be separated in analytically pure form by semipreparative
HPLC (see the ESI for details).