Angewandte
Communications
Chemie
Lowered reaction temperatures served only to decrease the
oxidant. Both HNO3[17] and FeCl3[18] are known to regenerate
rate of reaction.
DDQ from its hydroquinone form; however, one goal of this
approach to the Nazarov reaction was avoidance of strong
Brønsted or Lewis acids. Guided by reports from Liu and
Floreancig[19] and Helquist and co-workers[20] describing the
use of MnO2 and Mn(OAc)3 respectively as terminal oxidants
for DDQ, we applied these reagents to the oxidation-initiated
Nazarov process. In theory, one equivalent of MnO2 or two
equivalents of Mn(OAc)3 should suffice for complete oxida-
tion of the ether substrate via regeneration of catalytic DDQ.
However, in practice a larger excess of terminal oxidant was
found to be necessary to consume 1a in an experimentally
convenient timeframe (Figure 2). While these reactions
Aromatic substitution on the styryl moieties had only
minor effects on the reaction (entries 11–14). Methoxy-
substituted substrates 1k and 1l underwent successful Naz-
arov reaction in 2.5 and 1.0 hours, respectively, while 4-
chlorophenyl bearing substrates 1m and 1n gave the expected
products in 8 hours. These results suggest that the DDQ
oxidation step is rate limiting. Regardless of ether substitution
or terminal arene, all Nazarov cyclization products 2a–
n possessed an exocyclic methylene moiety resulting from
exclusive elimination with complete regioselectivity. Finally,
the presence of an unprotected alcohol was not tolerated, as
only ketone oxidation product 3a was obtained (entry 15).
We turned to unsymmetrical substrates to probe the
regioselectivity of the eliminative step (Scheme 3). Substrates
Figure 2. The use of catalytic DDQ with a MnIV or MnIII terminal
oxidant.
occurred in comparable yield to those using stoichiometric
quantities of DDQ, the time to completion was noticeably
greater, indicating that other co-oxidants may merit exami-
nation.
We have demonstrated the first oxidation-initiated Naz-
arov reactions of pentadienyl ethers, using DDQ to convert
a wide range of 3-alkoxy-1,4-pentadienes to pentadienyl
cations subject to standard Nazarov 4p electrocyclization.
The products are enol ethers bearing an exo methylene, the
result of a highly regioselective elimination step. Notably, the
exo-selective elimination leaves the two stereocenters gen-
erated in the electrocyclization intact. The oxidative Nazarov
cyclization permits the use of a new class of alternative
substrates, a mild method of activation, affords a new type of
dienol ether Nazarov products, and can be carried out with
substoichiometric quantities of the quinone oxidant.
Scheme 3. DDQ-initiated Nazarov cyclization products derived from
unsymmetrical substrates.
1p and 1q differentially substituted with aryl and isopropyl
groups at the termini underwent oxidative Nazarov cycliza-
tion in high yield to afford 2p and a pair of regioisomeric
dienol ethers 2q and iso-2q.[15] In both cases, high or complete
selectivity for elimination at the methyl group adjacent to the
isopropyl substituent was observed, which may arise from
lower steric demand by the isopropyl group.
Experimental Section
Representative procedure: Conversion of 1a to 2a via oxidative
Nazarov cyclization. Bis(allylic) ether 1a (56 mg, 0.20 mmol) was
dissolved in CH2Cl2, 2,3-dichloro-4,5-dicyano-1,4-benzoquinone
(55 mg, 0.24 mmol) was added, and the reaction mixture immediately
turned a deep purple color. The reaction was stirred at RT until
starting material was consumed as observed by TLC (4 h). The
reaction mixture was then filtered through a silica gel plug and eluted
with CH2Cl2 (50 mL). The filtrate was concentrated, and further
purification by flash chromatography gave 51 mg (92%) of enol ether
2a as a colorless oil.
Unsymmetrical substitution at the 2- and 4-positions of
the pentadienyl scaffold was also probed. Complete removal
of one methyl group (substrates 1r and 1s) did not affect the
overall efficiency of the oxidative Nazarov cyclization. More-
over, no evidence of competing endocyclic elimination was
observed, and only exo-methylene products 2r and 2s were
isolated. Substrates 1t and 1u, bearing methyl and n-propyl at
the 2- and 4-positions, showed high selectivity for elimination
from the more sterically accessible methyl group, affording 2t
exclusively and a 10:1 ratio of 2u and iso-2u as an inseparable
mixture.
Rf 0.51 (10:1 Hexane : EtOAc); IR (film): 3084, 3061, 3027, 2922,
1
2851, 1634, 1492, 1453, 1140 cmÀ1; H NMR (500 MHz, CDCl3): d =
7.32–7.27 (m, 4H), 7.25–7.20 (m, 2H), 7.14–7.09 (m, 4H), 5.08 (m,
1H), 4.54 (m, 1H), 3.87 (s, 3H), 3.66 (m, 1H), 3.57 (m, 1H), 1.67 ppm
(m, 3H); 13C NMR (125 MHz, CDCl3): d = 154.4, 150.7, 144.9, 143.9,
129.2, 128.6, 128.5, 127.8, 127.6, 126.6, 126.4, 101.6, 60.4, 59.7, 56.2,
11.9 ppm; HRMS calcꢀd for C20H20O (M+) 276.1514; found 276.1512.
The moderate toxicity of DDQ[16] and the potential for
HCN liberation in the presence of moisture prompted us to
consider use of catalytic DDQ with a stoichiometric terminal
Angew. Chem. Int. Ed. 2017, 56, 1 – 5
ꢀ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!