.
Angewandte
Communications
Table 3: (Continued)
Entry 13
N-tosylhydrazone 2H-naphthalenone Yield [%][a]
quant.
14
13n
d.r. 55:45[d]
15
13o
40[f]
[a] Yield of 14 isolated after neutral alumina chromatography, only
product obtained. [b] Reaction performed in PhH. [c] Determined using
1H NMR spectroscopy. [d] Determined using SFC-MS. [e] Identity of the
major diastereomer determined by nOe spectroscopic experiments.
[f] Styrene elimination product formed in 23%. Ac=acetyl, Boc=tert-
butyloxycarbonyl.
Scheme 2. Possible mechanisms for 2H-naphthalenone formation.
natively, 18 could be formed through a noncarbene mecha-
nism.[22] In this mechanism, the copper catalyst activates the
ester for attack by 15 to produce 22.[23] Loss of N2 forms the
benzyl cation 23, which is attacked by the copper alkoxide to
produce 18. Ring-opening of the epoxide generates the
oxocarbenium ion 19, which triggers a [1,2] methyl shift to
afford 20.[16] Dissociation of the copper catalyst produces the
product 12a.
Several experiments were performed in an attempt to gain
insight into the reactivity of the proposed catalytic inter-
mediates.[19] First, the addition of TEMPO (2,2,6,6-tetrame-
thylpiperidine 1-oxyl) to the reaction mixture did not produce
any new products to indicate that radical intermediates, if
formed, did not escape the solvent sheath. Attempts to trap
either 17 or 19 with a dipolarophile (DMAD; dimethylace-
tylenedicarboxylate)[4] or an external nucleophile (Et3SiH)
did not change the reaction outcome: only the 2H-naphtha-
lenone 12 was formed.[19] Next, we surveyed the reactivity of
the N-tosylhydrazones 24 to determine if the o-alkenyl group
was required for a successful reaction (Scheme 3). Irrespec-
tive of the potential ring formed, only b-hydride elimination
was observed. Finally, to examine the potential for an alkoxy-
substituted epoxide to rearrange, 27 was synthesized and its
reactivity toward Lewis and Brønsted acids was tested. We
found that while it remained inert toward mixtures of CuI and
the hydrazone R1 group. When the aldehyde-derived 13c was
submitted to the reaction conditions, the naphthalenol 14c
was produced (Table 3, entry 3). Next, the composition of the
ortho-alkenyl substituent was modified (entries 4–12). While
shrinking the ring-size led to an attenuated yield (entry 4),
substrates containing medium-sized o-cycloalkenyl rings were
smoothly converted into the a-methoxy-2H-naphthalenones
14e and 14 f (entries 5 and 6). The reaction also tolerated o-
heterocyclic substituents to afford 14g–i (entries 7–9). This
reaction does tolerate acyclic o-alkenyl groups: the hydrazone
13j was efficiently converted into the 2H-naphthalenone 14j
albeit with a reduced yield (entry 10). The diastereoselectivity
of the reaction was next probed (entries 11–14). While
exposure of the hydrazone 13k to the reaction conditions
provided a 3:2 mixture of diastereomers, the chiral, non-
racemic o-norbornyl 13l afforded the 2H-naphthalenone
product as a single diastereomer. Next, the effect of moving
the one of the bridge attachments from the allylic to the
homoallylic position was examined with the b-pinene-derived
13m (entry 13). Alleviating the steric pressure around the
carboxylate substituent improved the yield of the trans-
formation without the loss of stereoselectivity to produce
14m in 71% as a single diastereomer. In contrast, no
diastereoselectivity was observed with the (À)-menthol-
derived carboxylate 13n (entry 14). This reaction could also
be extended to the hydrazone 13o, which contained an o-aryl
group, to produce 14o in 40% (entry 15). Elimination,
however, also occurred to form the 2-substituted styrene in
23%.[19]
The conversion of the N-tosylhydrazones into a-methoxy-
2H-naphthalenones is proposed to occur through the tandem
cycloaddition/[1,2] rearrangement of
a copper carbene
(Scheme 2). The reaction of N-tosylhydrazone with tert-
butoxide generates the diazo 15,[20] which is decomposed by
the copper(I) salt to produce the copper carbene 16.[21] The
epoxide 18 could be produced through either a stepwise
reaction with the proximal methyl carboxylate via the ylide 17
or directly through a [2+1] cycloaddition reaction.[12] Alter-
Scheme 3. Experiments to elucidate the mechanism of the tandem
reaction.
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2015, 54, 12942 –12946