Angewandte
Chemie
desired but-3-ynyl side chain and efficiently delivered ketone
4 in 28% overall yield from the starting materials.
With a viable route for the synthesis of ketone 14 in hand,
1
the remaining challenge toward the preparation of the key
Conia-ene precursor 3 was the introduction of the proper
oxidation pattern in the northern quadrant of the molecule.
To address this task, ketone 14 was first transformed into the
corresponding enone by means of the Mukaiyama-dehydro-
genation procedure, furnishing the targeted compound in
[
12]
9
4% yield. Subsequent treatment with TBSOTf and 2,6-
lutidine in dichloromethane then selectively delivered silyl-
oxydiene 16. Inspired by a single preliminary report involving
an 1-acetoxy diene, we found that subjection of dienol silane
1
6 to an excess of CrO ·3,5-dimethylpyrazole complex in
3
dichloromethane directly led to the formation of the enedione
[13]
1
7 in a single step (66% overall yield for steps j and k).
The chlorination of enedione 17 at the a-position of the 5-
membered ring ketone turned out to be surprisingly difficult.
Selective formation of the lithium enolate at the sterically less
hindered C2-position of the molecule and subsequent treat-
ment with TsCl preferentially gave the corresponding tosyl
enol ether in 60% yield, instead of the chlorinated product,
which is usually observed under these conditions. Treat-
ment with TfCl, on the other hand, led to complex mixtures of
products. After quite some experimentation with many other
chlorinating agents, we eventually found that silyl enol ether
Scheme 3. Reagents and conditions: a) Me AlCl (20 mol%), toluene/
2
CH Cl (2:1), À158C, 24 h; then 08C, 24 h, 69%; b) F CCO H, CH Cl ,
2
2
3
2
2
2
RT, 2 h, 89%; c) 2-methyl-2-ethyl-1,3-dioxolane, TsOH·H O (25 mol%),
2
ethylene glycol (12 mol%), RT, 45 min, 95%; d) phosphazene base
P tBu (3.0 equiv), F C SO F (1.3 equiv), DMF, À108C to RT, 2 h; then
2
9
4
2
[
14]
RT for 24 h, 92%; e) nBuLi, THF, À788C to RT, 40 min; then BF ·OEt ,
3
2
À788C, 10 min; then DMF (3.3 equiv), 60 min, 78%; f) Pd/C
10 mol%), H (1 atm), EtOAc, RT, 2 h, 84%; g) dimethyl-1-diazo-2-
(
2
oxopropylphosphonate, K CO , MeOH, 2.5 h, 88%; h) nBuLi, Me SiCl,
2
3
3
THF, À788C to RT; then RT, 20 h; then 0.1m aq. HCl, THF, reflux, 4 h,
2%; i) LiN(SiMe ) , THF, À788C; then PhS(=N-tBu)Cl, 2 h, 94%;
1
8, upon exposure to Bu NCl (Mioskowskiꢀs reagent), was
4 3
9
3
2
directly transformed into a-chlorinated silyl enol ether 3,
j) TBSOTf, 2,6-lutidine, CH Cl , 08C, 1.5 h; k) CrO (15 equiv), 3,5-
2
2
3
[15]
without loss of the silyl group. This welcome result renders
dimethylpyrazole (15 equiv), CH Cl , À208C, 20 min, 66% for 2 steps.
2
2
unnecessary steps that would otherwise be required to
regenerate the silyl enol ether (Scheme 4).
Ts =toluene-4-sulfonyl, DMF=N,N-dimethylformamide, TBS=tert-
butyldimethylsilyl.
ketone in 9 was relatively inert towards attack by nucleo-
philes. Consequently, we opted to transform the methyl
ketone into the corresponding terminal acetylene via an
elimination reaction. To this end, two well-established
procedures were attempted in which the ketone is first
converted into its enol phosphate or enol triflate, before it is
treated, in a second step, with amide base to effect elimi-
[
8]
nation. However, after some unsuccessful experimentation
with the above protocols, a novel one-step procedure was
examined that had been reported to allow for the direct
conversion of methyl ketones into simple, terminal alkynes.
Scheme 4. Reagents and conditions: a) KN(SiMe ) (1.3 equiv), TBSCl
3
2
(
2.0 equiv), THF, À788C to RT, 3 h; b) Bu NCl (2.4 equiv), CH Cl ,
4
3
2
2
Treatment of methyl ketone 9 with phosphazene base P tBu
2
À788C to RT, 5 h, 51% for 2 steps; c) (MeCN)[(2-biphenyl)di-tert-
(
(
Schwesingerꢀs base) and nonafluorobutanesulfonyl fluoride
butylphosphine]gold(I) hexafluoroantimonate (0.5 equiv), acetone,
NfF) in dry DMF cleanly afforded terminal alkyne 10 in 92%
4
58C, 6 h, 65%; d) HCl (gas), SnCl (30 equiv), CH Cl , sealed tube,
4
2
2
[
9]
yield in a single step. Lewis acid promoted formylation of
the corresponding lithium acetylide with DMF followed by
the reduction of the intermediate propiolaldehyde with
catalytic Pd/C under an atmosphere of hydrogen then
À788C to RT, 5 h, 67%.
We then turned our attention to the crucial Conia-ene
reaction. After a few preliminary experiments with related
precursors (e.g. 18) and some minor optimization, we were
pleased to observe that treatment of 3 with acetonitrile[(2-
biphenyl)di-tert-butylphospine]gold(I) hexafluoroantimonate
(Echavarrenꢀs catalyst) in dry acetone at 458C effected both
the desired cyclization as well as the concomitant removal of
[
10]
furnished aliphatic aldehyde 12.
Transformation of the
aldehyde into the terminal acetylene 13 could subsequently
be brought about by means of the Ohira–Bestmann proce-
[
11]
dure. Finally, the sequence was completed by a one-pot
reaction, which included the protection of the terminal
acetylene with trimethylsilyl chloride and subsequent cleav-
age of the 1,3-dioxolane by using aqueous HCl. The sequence
described above reliably allowed for the introduction of the
[
16]
the silyl protecting group on the alkyne. It is worth noting
that although haloalkynes have been studied in gold-medi-
Angew. Chem. Int. Ed. 2012, 51, 13066 –13069
ꢀ 2012 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim