Nickel catalyzed Grob fragmentation: ω-dienyl aldehydes synthesis

Article abstract of DOI:10.1039/b610164j

With a proper choice of phosphane ligands, a Ni(cod)₂-phosphane catalyst promotes decarboxylative ring-opening reaction of a wide structural variety of cyclic carbonates 1 to give ω-dienyl aldehydes 2 in good yields. The Royal Society of Chemis

Full text of DOI:10.1039/b610164j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/chemcomm | ChemComm
Nickel catalyzed Grob fragmentation: v-dienyl aldehydes synthesis{
Masahiko Mori,b Masanari Kimura,b Yushi Takahashib and Yoshinao Tamaru*a
Received (in Cambridge, UK) 17th July 2006, Accepted 27th July 2006
First published as an Advance Article on the web 1st September 2006
DOI: 10.1039/b610164j
the isolation and X-ray structure characterization of I (without a
tether) prepared by the stoichiometric reaction of Ni(cod)2 [cod =
1,5-cyclooctadiene], a diene and an aldehyde in the presence of a
monodentate phosphane ligand.8
With a proper choice of phosphane ligands, a Ni(cod)2–
phosphane catalyst promotes decarboxylative ring-opening
reaction of a wide structural variety of cyclic carbonates 1 to
give v-dienyl aldehydes 2 in good yields.
These observations suggest that the equilibrium between I and 2
and M0 is strongly affected by the nature of the transition metal;
nickel tends to shift the equilibrium to the side of I, while
palladium works oppositely to drive the equilibrium to the side of
2 and Pd0. This contrasting behaviour between nickel and
palladium might be rationalized in part by the fact that nickel
forms a stronger metal–oxygen bond than palladium, e.g.,
DHu(Ni–O) = 60.5 kcal mol21 and DHu(Pd–O) = 58.9 kcal mol21
for cis-(PH3)2M(OMe)Me.9
The Grob fragmentation reaction is a very powerful tool for the
construction of desired molecules.1 However, harsh reaction
conditions, either strongly basic or acidic and/or high tempera-
tures, have limited its wide use.
In the last decade, C–C bond cleavage reactions catalyzed or
mediated by transition metals have gained significant attention,2,3
because they can be performed under much milder reaction
conditions and provide new organic transformations that are
otherwise difficult to achieve.
Here we disclose that Ni(cod)2, despite the above-mentioned
unfavorable reaction features, nicely catalyzes the decarboxylative
fragmentation reaction of a wide structural variety of 1 and
furnishes v-dienyl aldehydes 2 in comparable or better yields than
palladium species. However, unlike the palladium catalysis, which
In 1997, we revealed that many palladium species, especially
Pd2[(dba)3]2?CHCl3 [dba = dibenzylideneacetone] in acetonitrile,
catalytically promoted the Grob type decarboxylative ring-opening
reaction of cyclic carbonates 1 into v-dienyl aldehydes 2
(Scheme 1).4 The reaction proceeds at room temperature and is
rationalized by a facile b-C elimination of an oxapalladacyclo-
pentane intermediate I (M = Pd2+). On the other hand, we have
demonstrated that, Ni(acac)2 in the presence of triethylborane5 or
diethylzinc6 catalytically promotes the reductive cyclization
(homoallylation) of 2 and provides bis-homoallyl alcohols 3 in
good yields (Scheme 1).7 This formally reverse reaction was
supposed to proceed through oxidative addition of a nickel(0)
species across the diene and aldehyde moieties of 2, followed by the
worked equally well irrespective of the structure types of 1,4
a
nickel species required a fine tuning of the reaction conditions,
especially an appropriate choice of phosphane ligands, depending
on the structure types of 1.
As summarized in Table 1, carbonates characterized by their
ring strain (1a–c) and/or torsional strain (1d,e) smoothly under-
went the fragmentation reaction in the presence of Ni(cod)2
(10 mol%) and PPh3 (40 mol%) in acetonitrile at room temperature
and furnished 2 in comparable yields or even in better yields than
those obtained by Pd2(dba)3?CHCl3.
reduction of an oxanickellacyclopentane intermediate I (M = Ni2+
with Et3B or Et2Zn. Recently, the intermediacy of I (M = Ni2+
)
)
In sharp contrast to these, as shown in run 1, Table 2, a strain-
free six-membered carbonate 1f was robust and remained
unchanged under usual conditions (at 25 uC) and even under
forcing conditions undertaken at a higher temperature (81 uC).
This was rather surprising in view of a smooth fragmentation of 1f
with Pd2[(dba)3]2?CHCl3 at room temperature yielding 2f in 77%
isolated yield (E : Z = 10 : 1, footnote b, Table 2). Neither the
monodentate trialkyl- nor triarylphosphane ligands were effective
(runs 1–3). Among bidentate phosphane ligands, only those with
bite angles of ca. 95–105u effectively promoted the fragmentation
reaction (runs 6 and 7, Table 2).
was supported by Ogoshi and Kurosawa, who have succeeded in
Similar results were obtained for other cyclohexane derivatives
1j (runs 7 and 8, Table 3) and 1k (runs 9 and 10); with Ni(cod)2/
PPh3, they remained intact even after long heating at the refluxing
temperature of acetonitrile, while with Ni(cod)2/DPPF, they
smoothly underwent the ring-opening reaction at room tempera-
ture and provided 2j and 2k in much better yields than those with
Pd2[(dba)3]2?CHCl3.
Scheme 1 Nickel (or palladium) catalyzed fragmentation of 1 to 2 and
cyclization of 2 to 3 via a common intermediate I (M = Ni or Pd).
aDepartment of Applied Chemistry, Faculty of Engineering, Nagasaki
University, 1-14 Bunkyo, Nagasaki, 852-8521, Japan.
E-mail: tamaru@net.nagasaki-u.ac.jp
bGraduate School of Science and Technology, Nagasaki University, 1-14
Bunkyo, Nagasaki, 852-8521, Japan
The large bite angle of DPPF (99–105u) might be suited to the
tetrahedral geometry of a Ni0 species III and not to a square planer
geometry of a Ni2+ species II (Scheme 2).10 This destabilization of
{ Electronic supplementary information (ESI) available: Experimental
procedures and analytical and spectroscopic data for 1a–m13 and 2a–m.
See DOI: 10.1039/b610164j
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 4303–4305 | 4303

View Article Online
Table 1 Synthesis of v-dienyl aldehydes 2 by Ni-catalyzed decarboxy-
lative ring-opening reaction of 1a
Table 3 Acceleration of the Ni-catalyzed fragmentation of 1 with
bidentate phosphane ligands with wide bite anglesa
Yieldb (%)
Yield (E/Z) (%)
Run 1
t/h Product 2
Ni
Pd
Run Phosphane/solventb T/uC, t/h
Ni
Pd
1
2
75 (1 : 2.4) 65 (1 : 4.4)c
1
2
PPh3/AN
DPPF/THF
81, 24
25, 45
NR
91 (only E)
2
6
78 (2 : 1)
87 (1.7 : 1)c
82 (only E)c
3
4
0.5
96 (only E) 70 (only E)d
3
4
PPh3/AN
DPPF/THF
81, 24
25, 3
NR
81 (only E,E) 93 (only E,E)c
24
97 (1 : 1)
91 (1 : 1)c
5
6
PPh3/AN
DPPF/ether
81, 24
25, 9
24 (only E)e
98 (only E)
55 (only E)d
5
24
85 (1 : 1)
85 (1 : 3)d
7
8
PPh3/AN
DPPF/THF
81, 24
25, 4
NR
84 (3 : 1)
42 (2 : 1)d
a
Reaction conditions: 1 (1 mmol), Ni(cod)2 (0.1 mmol) and PPh3
(0.4 mmol) in acetonitrile (5 mL) at room temperature under N2. E/Z
b
c
Ratio of products in parenthesis.
Yield obtained using
Pd2(dba)3?CHCl3.4a d Ref. 4a. e A mixture of stereoisomers (see ESI).
9
10
PPh3/AN
DPPF/THF
81, 30
25, 3
NR
94 (4 : 1)
60 (1 : 1)d
II and stabilization of III by bidentate ligands with wide bite angles
might cooperate to facilitate the fragmentation of II and shift the
Table 2 Effects of phosphanes on the fragmentation of 1f
11
12
PPh3/AN
DPPF/Et2O
81, 15
25, 2
40 (1 : 5)
94 (2 : 1)
54 (1 : 4)c
Phosphane/
solventc
Yield
(E : Z)d (%)
13
14
15
a
PPh3/AN
DPPF/THF
DPPPent/THF
81, 40
25, 72
25, 30
NR
43 (only E)
Run
T/uC, t/h
Bite angle/u
100 (only E) 63 (only E)c
1
2
3
4
5
6
7
a
PPh3/AN
n-Bu3P/AN
Cy3P/AN
DPPE/THF
DPPP/THF
DPPB/THF
DPPF/THF
81, 24
81, 24
81, 24
25, 48
25, 48
25, 4
NR
NR
NR
NR
NR
76 (.20 : 1)
78 (18 : 1)
—
—
—
85
90–95
94–99
99–105
Reaction conditions:
1 (1 mmol), Ni(cod)2 (0.1 mmol) and a
phosphane (0.4 mmol for a monodentate and 0.2 mmol for a
bidentate ligand) in a solvent (5 mL) under N2. DPPPent = 1,5-
b
c
bis(diphenylphosphino)pentane. Ref. 4a. Yield obtained using
d
Pd2(dba)3?CHCl3.4a 59% conversion (59% isolated yield based on
conversion). A mixture of stereoisomers (see ESI).
e
f
25, 4
Reaction conditions: 1f (1 mmol, a mixture of stereoisomers, see
ESI), Ni(cod)2 (0.1 mmol) and phosphane (0.4 mmol for
monodentate and 0.2 mmol for a bidentate ligand) in a solvent
equilibrium to the side of a mixture of 2 and III. The complex III,
however, might be reactive enough toward oxidative addition to
allylic carbonates 1 and might be able to supply II, hence
accomplishing the catalytic cycle.
a
a
b
(5 mL) under N2. 2f in 77% (E : Z = 10 : 1) using Pd2(dba)3?
CHCl3 as a catalyst.4a AN = acetonitrile, THF = tetrahydrofuran,
c
Cy = cyclohexyl, DPPE = 1,2-bis(diphenylphosphino)ethane, DPPP
= 1,3-bis(diphenylphosphino)propane, DPPB = 1,4-bis(diphenylpho-
sphino)butane, DPPF
Yields refer to the isolated spectroscopically homogeneous 2f.
In Table 3 are summarized the carbonates 1 that are either
unreactive (NR) or provide 2 albeit in low yields with the catalyst
Ni(cod)2/PPh3. Comparison of the data in Tables 1 and 3 clearly
=
1,19-bis(diphenylphosphino)ferrocene.
d
4304 | Chem. Commun., 2006, 4303–4305
This journal is ß The Royal Society of Chemistry 2006

View Article Online
structural variety of cyclic carbonates 1 to furnish v-dienyl
aldehydes 2, where the phosphane ligands play a pivotal role to
shift the equilibrium between an intermediate I and an v-dienyl
aldehyde 2 to the product side. The reaction shows better yields
and higher (E)-selectivity than the original palladium-catalyzed
reaction reported previously by the authors.4a The ready
availability of the starting materials 1 and the importance of the
products 2 as strategic intermediates for natural product synthesis
may augment the utility of the present reaction as a synthetic
method. Furthermore, the reaction provides examples that
explicitly demonstrate the salient difference in catalytic reactivity
between a nickel(0) and a palladium(0) species.
Scheme 2 Acceleration of b-C elimination by bidentate ligands with
wide bite angles (h).
Financial support from the Ministry of Education, Culture,
Sports, Science and Technology, Japanese Government [Grant-in-
Aid for Scientific Research (B) 16350058 and Priority Areas
17035065] is gratefully acknowledged.
Scheme 3 Increase in steric repulsion between P1 and vinyl moiety in a
transition state V, placing all the P1, P2, Ni, C5, C4, C3 and O atoms in
the same plane.
indicates that both the C19- and C29-substituents on the vinyl
group of 1 seriously retard the reaction. For example, while 1b
undergoes fragmentation at room temperature with Ni(cod)2/PPh3
(run 2, Table 1), the C19-Me and C29-Me derivatives, 1g and 1h,
remain unchanged under the same or even under forcing
conditions (runs 1 and 3, Table 3). The fragmentation of 1g and
1h was successfully achieved by making use of the Ni(cod)2/DPPF
catalytic system (runs 2 and 4). More contrasting results were
observed between the reactions of 1e (run 5, Table 1) and 1m (runs
13–15). For the latter substrate, even the Ni(cod)2/DPPF catalytic
system was not effective enough, and Ni(cod)2/DPPPent [1,5-
bis(diphenylphosphino)pentane] turned out to work much better.11
The retardation of the reaction by the C19 and C29 substituents,
especially by the one on C19 (cf, the reaction times in runs 2 and 4,
Table 3), might be ascribed to the increase in the steric repulsion
between P1 and the vinyl group moiety that arises when IV
approaches a transition state V (Scheme 3), since this process is
accompanied by the decrease in the absolute value of the dihedral
angle Ni–C5–C4–C3, and hence the decrease in the dihedral angle
P1–Ni–C5–C19. That is, there seems to exist an interesting
dichotomy in the role of bidentate phosphane ligands with wide
bite angles; electronically they accelerate the fragmentation
reaction by stabilizing III and destabilizing II (Scheme 2),12 while
sterically they retard the reaction preventing the Ni–C5 bond from
taking coplanar conformation with the C3–C4 bond.
Notes and references
1 G. Mehta and R. S. Kumaran, Tetrahedron Lett., 2005, 46, 8831–8835;
T. Constantieux and J. Rodriguez, Sci. Synth., 2004, 26, 413–462, and
references therein.
2 b-C Elimination of strained cyclobutanes and cyclopropanes:
M. Murakami, S. Ashida and T. Matsuda, J. Am. Chem. Soc., 2006,
128, 2166–2167; M. Murakami, S. Ashida and T. Matsuda, J. Am.
Chem. Soc., 2005, 127, 6932–6933; T. Nishimura and S. Uemura,
Synlett, 2004, 201–216, and references therein.
3 Norbornene-mediated reactions: C. Bressy, D. Alberico and M. Lautens,
J. Am. Chem. Soc., 2005, 127, 13148–13149; M. Catellani, E. Motti,
F. Faccini and R. Ferraccioli, Pure Appl. Chem., 2005, 77, 1243–1248.
4 (a) H. Harayama, T. Kuroki, M. Kimura, S. Tanaka and Y. Tamaru,
Angew. Chem., Int. Ed. Engl., 1997, 36, 2352–2354; see also:
H. Harayama, M. Kimura, S. Tanaka and Y. Tamaru, Tetrahedron
Lett., 1998, 39, 8475–8478.
5 M. Kimura, A. Ezoe, K. Shibata and Y. Tamaru, J. Am. Chem. Soc.,
1998, 120, 4033–4034; see also: M. Kimura, A. Ezoe, S. Tanaka and
Y. Tamaru, Angew. Chem., Int. Ed., 2001, 40, 3600–3602; K. Shibata,
M. Kimura, M. Shimizu and Y. Tamaru, Org. Lett., 2001, 3,
2181–2183.
6 M. Kimura, A. Miyachi, K. Kojima, S. Tanaka and Y. Tamaru, J. Am.
Chem. Soc., 2004, 126, 14360–14361; M. Kimura, H. Fujimatsu,
A. Ezoe, K. Shibata, M. Shimizu, S. Matsumoto and Y. Tamaru,
Angew. Chem., Int. Ed., 1999, 38, 397–400.
7 Closely related reactions involving I as an intermediate: M. Kimura,
A. Ezoe, M. Mori and Y. Tamaru, J. Am. Chem. Soc., 2005, 127,
201–209; A. Ezoe, M. Kimura, I. Inoue, M. Mori and Y. Tamaru,
Angew. Chem., Int. Ed., 2002, 41, 2784–2786; M. Kimura, S. Matsuo,
K. Shibata and Y. Tamaru, Angew. Chem., Int. Ed., 1999, 38,
3386–3388.
8 S. Ogoshi, K. Tonomori, M. Oka and H. Kurosawa, J. Am. Chem.
Soc., 2006, 128, 7077–7086.
9 S. A. Macgregor and G. W. Neave, Organometallics, 2004, 23, 891–899.
10 E. Burello, P. Marion, J.-C. Galland, A. Chamard and G. Rothenberg,
Adv. Synth. Catal., 2005, 347, 803–810.
Cyclic carbonates 1a–m were prepared readily according to the
following three-step procedures (see, ESI{): (1) cross-aldol of
a,b-unsaturated aldehyde with lithium enolate of cyclic ketone, (2)
LiAlH4 reduction to diol, (3) cyclic carbonation with methyl
chloroformate/triethylamine.
11 Bite angles (103.55–103.86u) of DPPPent for CuI complexes determined
by X-ray crystallography: Effendy, C. D. Nicola, M. Fianchini,
C. Pettinari, B. W. Skelton, N. Somers and A. H. White, Inorg.
Chim. Acta, 2005, 358, 763–795; bite angle (104.37u) for a palladium
complex obtained by DFT-optimization: R. Kuwano and M. Yokogi,
Org. Lett., 2005, 7, 945–947.
12 Three different fragmentation modes of nickellacyclopentanes:
R. H. Grubbs, A. Miyashita, M. Liu and P. Burk, J. Am. Chem.
Soc., 1978, 100, 2418–2425.
13 Crystal data for 1c: C14H20O3, M = 236.31, orthorhombic, space group
P212121, a = 7.5620(4), b = 10.2674(8), c = 16.5730(11) s, U =
1286.76(15) s3, T = 296.1 K, Z = 4, m(Mo-Ka) = 0.7107 mm21, 1551
reflections measured, 985 unique (Rint = 0.025), wR(F2) = 0.1596 (all
data). For crystallographic data in CIF or other electronic format see
DOI: 10.1039/b610164j.
Finally, it may be worth noting that the present nickel catalyzed
degradation not only records better yields, but also shows higher
(E)-selectivity than the palladium catalyzed reaction. For example,
a mixture of (E)- and (Z)-2k was obtained in a ratio of 1 : 1 in 60%
isolated yield under the palladium catalysis, while under the nickel
catalysis the same product was obtained in a ratio of 4 : 1 in 94%
yield (run 10, Table 3). More contrasting results were observed for
the reaction of 1l, where the yield was doubled and the
stereoselectivity was reversed from the (Z)-selective one to the
(E)-selective one (run 12, Table 3).
In summary, the Ni(cod)2/PPh3 or Ni(cod)2/DPPF catalytic
system promotes a novel decarboxylative fragmentation of a wide
This journal is ß The Royal Society of Chemistry 2006
Chem. Commun., 2006, 4303–4305 | 4305