Optimization of a Pd-catalyzed intramolecular α-arylation synthesis of tricyclo-[7.3.1.02,7]-trideca-2,4,6-trien-13-ones

Article abstract of DOI:10.1016/j.tetlet.2010.06.085

We have optimized the Pd-catalyzed intramolecular α-arylation of 2-(2-halo-benzyl)-cyclohexanones and found that the use of 2- (dicyclohexylphosphino)-2′,4′,6′-tri-i-propyl-1, 1′-biphenyl (X-Phos) as an added ligand led to a reproducible, efficient, and scalable synthesis of tricycle-[7.3.1.0^2,7]-trideca-2,4,6-trien- 13-ones.

Full text of DOI:10.1016/j.tetlet.2010.06.085

Tetrahedron Letters 51 (2010) 4441–4444
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Optimization of a Pd-catalyzed intramolecular
a-arylation synthesis
of tricyclo-[7.3.1.02,7]-trideca-2,4,6-trien-13-ones
à
§
–
*
Noel A. Powell , Timothy J. Hagen , Fred L. Ciske , Cuiman Cai , Joseph E. Duran, Daniel D. Holsworth ,
Daniele Leonard à, Robert M. Kennedy ||, Jeremy J. Edmunds
Department of Chemistry, Michigan Laboratories, Pfizer Global Research & Development, Ann Arbor, MI, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
We have optimized the Pd-catalyzed intramolecular
a-arylation of 2-(2-halo-benzyl)-cyclohexanones
and found that the use of 2-(dicyclohexylphosphino)-20,40,60-tri-i-propyl-1,10-biphenyl (X-Phos) as an
added ligand led to a reproducible, efficient, and scalable synthesis of tricycle-[7.3.1.02,7]-trideca-2,4,6-
trien-13-ones.
Received 18 February 2010
Revised 15 June 2010
Accepted 16 June 2010
Available online 20 June 2010
Ó 2010 Elsevier Ltd. All rights reserved.
Structurally complex polycyclic structures have long inspired
synthetic chemists due to their presence in a wide variety of biolog-
ically active natural products. In the pharmaceutical industry, poly-
cyclic structures are often targeted during the drug discovery
process as a strategy to improve the potency and/or pharmacoki-
NH
O
O
N
O
CHO
CH2Cl2
MeOH, r.t.
~100%
-38 to 0 °C
77%
2
1
netic properties of
a series of analogs through reduced
conformational flexibility. We were interested in the tricyclo-
[7.3.1.02,7]-trideca-2,4,6-trien-13-one 6 as a key intermediate for a
medicinal chemistry program and required a concise synthetic route
capable of producing multi-gram quantities for further elaboration.
Our first-generation route (Scheme 1) involved a lengthy synthetic
sequence starting with the b-tetralone 1.1 While >50 g of 6 could
be prepared via this route, we recognized that this route was labori-
ous, inefficient, and not readily amenable for the rapid investigation
of the structure–activity relationships (SARs) around the tricyclic
core ring structure. Consequently, we sought a route that would be
shorter in length, more atom efficient, and readily amenable to
SAR studies around the tricyclic core.
LiBr
TsCl
Li2CO3
OH
OTs
Pyridine
DMA
reflux
~100%
36 °C 5 days
3
4
95%
H2, Pd/C
O
O
EtOH
71%
5
6
Scheme 1. First generation synthesis of the tricycle-[7.3.1.02,7]-tridec-2,4,6-trien-
Muratake et al. have reported a concise synthesis of 6 through a
13-one core.
Pd-catalyzed intramolecular ketone a-arylation of 2-(2-bromoben-
zyl)-cyclohexanone 7 (Fig. 1).2 Since this route represented a sub-
stantial improvement over the original synthesis, we attempted to
O
* Corresponding author at present address: Biogen Idec, 14 Cambridge Center,
Cambridge, MA 02142, USA. Tel.: +1 617 914 1152; fax: +1 617 914 4885.
E-mail addresses: napowell@comcast.net, noel.powell@biogenidec.com (N.A.
Powell).
THF, 100 °C
sealed tube
6
Br
83%
Present address: Northern Illinois University, DeKalb, IL 60115, USA.
Present address: Cayman Chemical, 1180 East Ellsworth Road, Ann Arbor, MI
à
10 mol% PdCl2(PPh3)2
3 equiv Cs2CO3
48108, USA.
§
Present address: Pfizer Global Research & Development, Groton Labs, Groton, CT,
O
no product
formation
7
USA.
THF
–
Present address: StemNext LLC, San Diego, CA 92129, USA.
Present address: Vestaron Corporation, 4717 Campus Drive, Kalamazoo, MI
reflux
||
49008, USA.
Figure 1. Initial attempts to realize intramolecular ketone
2-(2-bromobenzyl)-cyclohexanone 7.
a-arylation of
Present address: Abbott Labs, 381 Plantation Street, Worchester, MA 01605, USA.
0040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.06.085

4442
N. A. Powell et al. / Tetrahedron Letters 51 (2010) 4441–4444
Table 2
repeat their results. Accordingly, we were pleased to find that the
treatment of 7 with 10 mol % PdCl2(PPh3)2 and 3 equiv of Cs2CO3 in
THF in a sealed tube at 100 °C afforded the desired tricyclic ketone
6 in 83% yield on a 50 mg scale. However, we were unsuccessful in
reproducing these results at reflux under atmospheric pressure con-
ditions. Since sealed tube conditions were unlikely to support the
synthesis of the multi-gram quantities of ketone 6 that would be re-
quired to support our medicinal chemistry efforts, we embarked on
Optimization of intramolecular
a
-arylation of ketone 3
Pd
O
Br
F
O
O
ligand
+
Base
solvent
100 °C
F
8
10
an optimization of the Pd-catalyzed intramolecular ketone
a-aryla-
9
F
tion of 7. Iwama and Rawal have recently reported the successful Pd-
catalyzed intramolecular
a-arylation of trimethysilyl enol ether
derivativesusinga Pd2(dba)3/t-Bu3P catalystsystem.3 While directly
applicable to the synthesis of tricyclo-[7.3.1.02,7]-trideca-2,4,6-tri-
en-13-ones, we sought to avoid the required formation of the tri-
methylsilyl enol ether and further investigated the cyclization of
the parent ketone.4,5 Recently, Khartulyari and Maier have reported
Entry
Pd catalysta
Ligandb
Basec
Solvent
Yield (%)
a
b
c
PdCl2(PPh3)2
PdCl2(PPh3)2
PdCl2(PPh3)2
—
—
—
Cs2CO3
Cs2CO3
NaOt-Bu
THF
PhCH3
PhCH3
39 9
NR
14 9,
13 10
34 9,
21 10
NR
53 9
50 9
63 9
32 9
NR
NR
NR
NR
40 9
67 9
84 9
68 9
NR
success in the intramolecular
a-arylation of ketones to prepare
d
PdCl2(PPh3)2
—
K3PO4
PhCH3
structurally similar benzomorphan derivatives in moderate yields
using the Pd(dba)2/t-Bu3P catalyst system.6 This report emerged
after our studies were completed, but indicates that others have also
succeeded in avoiding the preliminary formation of silyl enol ethers.
Initial reaction optimization studies were conducted with 2-(2-
bromobenzyl)-cyclohexanone 7 (Table 1).7 Interestingly, heating
the reaction mixture in the non-polar solvent toluene at reflux gave
no detectable product (entry a), which was at odds with the results
reported by Muratake et al.2 However, switching to a more polar sol-
vent, such as DMF, did afford the desired product 6 in low yield when
heated at 80 °C (entry b). Encouragingly, higher reaction tempera-
tures led to a greatly improved 72% yield of 6 (entry c). Other polar
solvents, such as 1,4-dioxane or DMA, were also successful in
e
f
g
h
i
j
k
l
m
n
o
p
q
r
PdCl2(PPh3)2
PdCl2(PPh3)2
PdCl2(PPh3)2
PdCl2(PPh3)2
Pd (PPh3)4
Pd(Pt-Bu3)2
PdCl2ꢀdppf
POPd
POPd2
Pd(OAc)2
Pd(OAc)2
Pd(OAc)2
—
—
—
—
—
Et3N
PhCH3
1,4-Dioxane
DME
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
DMF
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
PhCH3
—
—
—
11
12
13
14
15
Pd(OAc)2
Pd(OAc)2
a
b
c
Pd catalyst loading of 10 mol %.
Ligand loading of 20 mol %.
3 equiv of base.
promoting the
a-arylation reaction in moderate to good yields
(48–72%, entries d and e). The use of DMF as a solvent proved to be
scalable, as a 16 mmol scale reaction afforded 6 in 66% yield (entry f).
While acceptable yields of 6 under non-sealed tube conditions
were obtained by simply using more polar solvents, such as DMF
and DMA, we were concerned about the isolation of 6 from large
quantities of DMF or DMA. Furthermore, the use of 3 equiv of
Cs2CO3 on multi-gram scale could be problematic and expensive.
Consequently, we further investigated the optimization of the
reaction conditions with the goal of identifying reaction conditions
utilizing a less expensive base and a more tractable solvent. Be-
cause the electron-rich phenyl ring present in 6 was viewed as a
potential metabolic liability, we chose to install an electron-with-
drawing fluorine substituent on the phenyl ring and utilize sub-
strate 8 in our optimization experiments. All reactions were
performed in a parallel fashion in a Radley reaction carousel (Table
2). While low yields of the desired cyclization product 9 were ob-
tained in refluxing THF with Cs2CO3 as a base (entry a); Cs2CO3 in
toluene resulted in no reaction (entry b). The use of more soluble
inorganic bases in toluene afforded low yields of 9 (entries c and
d), while amine bases completely inhibited the reaction (entry e).
The des-bromo compound 10 was identified as the major by-prod-
uct in these reactions, presumably as a result of competitive pro-
teo-reductive elimination in the catalytic cycle. A solvent survey
with K3PO4 as the base indicated more polar solvents, such as
1,4-dioxane, DME, and DMF, afforded higher yields of 9 (entries
f–h). A catalyst survey determined that Pd(PPh3)4 and PdCl2(PPh3)2
gave equal yields of product, while Pd(t-Bu3P)2, PdCl2(dppf), and
the cationic POPd and POPd2 catalysts8 gave no reaction (entries
i–m). The failure of Pd(t-Bu3P)2 to catalyze the reaction is signifi-
cant, given the successful use of in situ-prepared Pd(t-Bu3P)2 by
Iwama & Rawal and Khartulyari & Maier in similar reactions.3,6
Table 1
Solvent affects on the Pd-catalyzed intramolecular ketone
a-arylation of substrate 7
PdCl2(PPh3)2
3 equiv Cs2CO3
Br
O
solvent, Δ
7
O
6
Entry
Scale (mmol)
Mol % Pd catalyst
Solvent
Temp (°C)
Yield (%)
a
b
c
d
e
f
10
19
1
1
1
20
8
10
10
10
10
Toluene
DMF
DMF
1,4-Dioxane
DMA
DMF
Reflux
80
115
Reflux
115
115
NRa
5
72
48
72
66
16
a
NR = no reaction.

N. A. Powell et al. / Tetrahedron Letters 51 (2010) 4441–4444
4443
potentially increase the rate of the desired C–C bond-forming
reductive elimination event and decrease the rate of the proteo-
reductive elimination event. Toward this end, we were encouraged
by the reports of Buchwald and co-workers on the use of electron-
PCy2
PCy2
PtBu2
rich phosphine ligands in intermolecular a
-arylations of ketones.9
11
JohnPhos
12
Screening of several (Fig. 2) phosphine ligands 11–15 (entries n–r)
with Pd(OAc)2 and K3PO4 in toluene identified X-Phos 13 as the
most successful ligand, yielding 9 in 84% yield with no detectable
amount of the des-bromo by-product 10 (entry p). We believe that
the success of the X-Phos 13 ligand is due to the highly sterically
encumbered environment about the P atom, which significantly
enhances the rate of the desired C–C bond reductive elimination
event. The reaction conditions of X-Phos 13/Pd(OAc)2 and K3PO4
in toluene met our goal to identify more tractable reaction condi-
tions for the conversion of 8 to 9, and have proven to be scalable
to >100 g scale. These conditions provided 9 as a racemic mixture
of a single diastereomer. We were interested in developing an
enantioselective version on this cyclization. However, initial efforts
13
X-Phos
PPh2
PPh2
O
PCy2
PCy2
14
Xantphos
15
rac-BINAP
Figure 2. Phosphine ligands used in optimization studies.
toward an enantioselective a-arylation of 8 were not successful, as
the use of rac-BINAP 15 as a ligand resulted in no reaction (entry r).
This precluded us from examining the separate BINAP enantio-
mers. However, it is possible that other chiral phosphine ligands
may promote the cyclization of 8 in an enantioselective fashion.
Once optimal, reproducible, and scalable reaction conditions
were identified, we examined the scope of these conditions with a
variety of 2-(2-halobenzyl)-cyclohexanones (Table 3). Electron-rich
substrates such as 16a underwent facile cyclization to yield 17a in
60% yield. Aryl chlorides were also useful cyclization substrates, as
16b and 16c afforded the corresponding cyclic products 17b and
17c in 61% and 84% yields, respectively. Heterocyclic chlorides were
also tolerated, although the cyclization of 16d led to product 17d in a
reduced26%yield. Variationwasalsotoleratedinthecyclohexanone
portion, as the 4-oxo-cyclohexanone substrates 16e and 16f under-
went cyclization to afford 17e and 17f in >98% yields. The 2,2-disub-
stituted cyclohexanone 16g also underwent cyclization to yield 17g
in good yield. Substrates such as 17g may also be useful to access
enantiomerically pure tricyclo-[7.3.1.02,7]-trideca-2,4,6-trien-13-
ones if the quaternary center in 16g can be prepared in an enantiose-
lectivefashion. Alternatively, thecarboxylatefunctionalgroupcould
be derivatized as a suitable chiral auxiliary that might provide suffi-
cient enolate facial selectivity.
Table 3
Scope of the intramolecular ketone
a
-arylation of 2-(2-halobenzyl)-cyclohexanones
Pd(OAc)2 (10 mole%)
X-Phos (20 mole%)
Z
Y
X
Z
Y
O
R
K3PO4, PhCH3, 100 °C
R
16
O
17
Entry
Substrate
Product
Yield (%)
60
Br
O
a
b
c
d
e
f
MeO
MeO
O
Cl
O
61
84
F3C
F
F3C
F
O
Cl
O
O
O
N
Cl
N
O
2. General bench-scale synthetic procedure for the
intramolecular a-arylation of 2-(2-halobenzyl)-cyclohexanones
using Pd(OAc)2/X-Phos
26
O
Br
O
O
O
O
O
98
2.1. Synthesis of 5-fluoro-tricyclo-[7.3.1.02,7]-trideca-2,4,6-
trien-13-one (9)
F
F
F
F
F
O
O
Br
Anoven-dried, 100 mL round-bottommedflaskwas cooledunder
Ar and charged with 140 mg (0.62 mmol) of Pd(OAc)2, 592 mg
(1.24 mmol) of X-Phos 13, and 3.03 g (14.3 mmol) of anhydrous
K3PO4. A solution of 1.77 g (6.21 mmol) of 2-(2-bromo-5-fluoroben-
zyl)-cyclohexanone 3 in 25 mL of anhydrous toluene was added, and
the resulting suspension was heated to reflux for 18 h. After cooling
to room temperature, the reaction mixture was diluted with EtOAc,
filtered through Celite, and concentrated. Purification by flash col-
umn chromatography (SiO2, 100% hexanes then gradient to 10%
EtOAc/hexanes) gave 1.067 g (84%) of 4 as a clear viscous oil. 1H
NMR (400 MHz, CDCl3) d 1.45–1.63 (m, 2H), 1.95–2.15 (m, 4H),
2.74–2.78 (m, 1H), 3.17 (d, J = 17.8 Hz, 1H), 3.43 (dd, J = 17.8,
7.2 Hz, 1H), 3.45–3.49 (m, 1H), 6.84–6.88 (m, 1H), 6.90–6.93 (m,
1H), 6.94 (dd, J = 12.0 Hz, 5.9 Hz, 1H); MS calcd for [C13H13FO+1]:
205.2. Found (APCI): 205.1 (M+1).
100
F
O
O
Br
g
74
F
EtO2C
F
CO2Et
Since proteo-reductive elimination to form the des-bromo
by-product 10 remained a significant issue, we investigated the
addition of sterically hindered electron-rich phosphine ligands to

4444
N. A. Powell et al. / Tetrahedron Letters 51 (2010) 4441–4444
3. General large-scale synthetic procedure
Acknowledgments
3.1. Synthesis of 5-fluoro-tricyclo-[7.3.1.02,7]-trideca-2,4,6-
trien-13-one (9)
The authors thank S. R. Damireddi, S. Kumar, Y. Huang, and S.
Reddy of PharmAgra Labs, www.pharmagra.com, for their assis-
tance in performing the large-scale synthesis of compound 4.
A 5-L multi-neck round-bottommed flask equipped with a stir
bar was charged with 575 g (2.06 mol) of 2-(2-bromo-5-fluoroben-
zyl)-cyclohexanone 3, 85 g (0.178 mol) of X-Phos 13, 990 g
(4.65 mol) of anhydrous K3PO4, and 5.5 L of toluene. The mixture
was degassed under vacuum and then treated with 42 g
(0.19 mol) of Pd(OAc)2. The reaction mixture was heated to reflux
for 36–40 h until complete by TLC. After cooling to room tempera-
ture, the reaction mixture was filtered through Celite, washed with
EtOAc, and concentrated to a dark orange oil. The crude material
was purified by chromatography on silica gel, eluting with 97:3
cyclohexane/EtOAc. An impurity with slightly higher Rf than the
product complicated the purification. The cleanest fractions were
combined and concentrated to give 185 g of 4 (45% yield, 96.8%
pure by GC). Concentration of impure fractions gave an additional
60 g of 4 that was 89% pure by GC for an overall yield of 59%.
Supplementary data
Supplementary data (experimental procedures and characteriza-
tion data for compounds 9 and 17a–g) associated with this article
can be found, in the online version, at doi:10.1016/j.tetlet.2010.
06.085.
References and notes
1. Mitsuhashi, K.; Shiotani, S. Chem. Pharm. Bull. 1970, 18, 75–87.
2. (a) Muratake, H.; Natsume, M.; Nakai, H. Tetrahedron 2004, 60, 11783–11803;
(b) Muratake, H.; Natsume, M. Tetrahedron Lett. 1997, 38, 7581–7582.
3. Iwama, T.; Rawal, V. H. Org. Lett. 2006, 8, 5725–5728.
4. For a related Heck approach to the tricyclo[7.3.1.02,7]trideca-2,4,6-triene ring
structure, see: Kelly, S. A.; Foricher, Y.; Mann, J.; Bentley, J. M. Org. Biomol. Chem.
2003, 1, 2865–2876.
5. For additional examples of intramolecular
a-arylations of carbonyls, see: (a)
Solé, D.; Vallverdú, L.; Solans, X.; Font-Bardia, M.; Bonjoch, J. J. Am. Chem. Soc.
2003, 125, 1587–1594; (b) Shaughnessy, K. H.; Hammann, B. C.; Hartwig, J. F. J.
Org. Chem. 1998, 63, 6546–6553; (c) Honda, T.; Sakamaki, Y. Tetrahedron Lett.
2005, 46, 6823–6825.
In summary, we have demonstrated that use of the X-Phos li-
gand 13 in the Pd-catalyzed intramolecular ketone
a-arylation of
2-(2-halobenzyl)-cyclohexanones results in an efficient, reproduc-
ible, and scalable synthesis of tricyclo-[7.3.1.02,7]-trideca-2,4,6-tri-
en-13-ones. A variety of substitution patterns on the aryl halide
and cyclohexanone portions are tolerated. Both aryl chlorides
and bromides undergo facile reaction. This route has enabled the
rapid investigation of structure–activity relationships about the
tricyclic core ring structure. We believe that the optimization con-
ditions detailed herein will be of interest to others investigating
6. Khartulyari, A. S.; Maier, M. E. Eur. J. Org. Chem. 2007, 317–324.
7. Prepared by alkylation of 1-pyrrolidino-1-cyclohexane with 2-bromobenzyl
bromide: Booth, S. E.; Jenkins, P. R.; Swain, C. J.; Sweeney, J. B. J. Chem. Soc.,
Perkin Trans. 1 1994, 23, 3499–3508.
8. Li, George Y.; Zheng, G.; Noonan, A. F. J. Org. Chem. 2001, 66, 8677–8681.
9. (a) Palucki, M.; Buchwald, S. L. J. Am. Chem. Soc. 1997, 119, 11108–11109; (b)
Fox, J. M.; Huang, X.; Chieffi, A.; Buchwald, S. L. J. Am. Chem. Soc. 2000, 122,
1360–1370.
the intramolecular a-arylation of carbonyl compounds.