2
M.S. Freire Franco et al. / Tetrahedron Letters xxx (2018) xxx–xxx
Fig. 1. Commercially available drugs and other biologically active compounds containing the quinoxaline core.
Scheme 1. Synthetic strategies to obtain 2-styryl quinoxalines.
Scheme 2. Palladium-catalyzed Fujiwara-Moritani reaction using quinoxaline N-oxide and styrenes and acrylates.
When other solvents were used, such as 2-propanol, acetoni-
ever, silver carbonate was found to be a better oxidant than silver
acetate for synthesizing compounds 3c and 3f (Scheme 3).
trile, DMSO, DMF, and toluene (Table S1, entries 2–6; ESI), no
improvement in yield was observed. Moderate yields were also
obtained when 1,4-dioxane was used with different sources of
Pd(II) (Table 1, entries 2–4). The use of bidentate ligands such
as 1,10-phenanthroline (Phen) and 4,4-di-tert-butyl-2,2-bipyridyl
BBBPY) were found to decrease the yields of the olefination pro-
duct (Table 1, entries 5 and 6) compared to the pyridine ligand
Table 1, entry 1). Subsequently, different amino acids were used
Quinoxaline N-oxide proved to be an ideal substrate for the
Fujiwara-Moritani coupling and presented specific reactivity of
the N-oxide functionality at the ortho position. The remarkable
ability of the mono-N-protected amino acid (Fmoc-Val-OH) to
accelerate Pd(II)-catalyzed oxidative C–H cross-coupling reactions,
first discovered by Yu and co-workers, was also observed in this
(
3
8,39
(
study.
as ligands (Table 1, entries 7–10). Among these, Fmoc-Val-OH
gave the best result (Table 1, entry 10). Several oxidants were
also screened in order to improve the yield (Table 1, entries
A probable reaction mechanism has been proposed (Scheme 4)
based on the experimental results obtained from this work and the
extensive experimental and computational studies described in the
3
8,40,41
1
0–14). Use of silver acetate led to the highest yield of (E)-2-
literature.
It is proposed that palladium(II) acetate initially
styrylquinoxaline N-oxide (Table 1, entry 14,). Upon investigating
the different ratios of quinoxaline N-oxide/styrene (Table 1,
entries 15–18), the 4:1 ratio was found to be optimal. Reduction
in the amount of the additive led to a considerable decrease in
the yield (Table 1, entries 19–21); in the absence of an additive,
a poor yield was obtained (Table 1, entry 22). Furthermore, no
reaction was possible without an oxidant or a palladium source
forms a complex with the N-protected amino acid, with both the
–NH and carboxylate groups coordinating with the Pd center in a
bidentate manner. Subsequently, this Pd-complex coordinates
with the oxygen at the N-oxide and reversible ligand exchange
takes place by concerted metalation/deprotonation (CMD), forming
complex C (Scheme 4), which is relatively stable due to the forma-
tion of a hydrogen-bond between the acetic acid ligand and the
anionic oxygen. That the coordination occurs with the N-oxide
moiety and not with nitrogen was verified by the experimental
observation that the neutral quinoxaline did not yield the alkeny-
lated product (result not showed). Finally, coordination of Pd to the
(Table 1, entries 23 and 24).
After establishing the optimal reaction conditions (Table 1,
entry 14,), the scope and limitations of the protocol were investi-
gated (Scheme 3). Typically, when silver acetate was used as the
oxidant, it was possible to obtain the desired product in good to
excellent yield, even with bromine-substituted styrene 3e. How-
styrene followed by carbopalladation leads to complex
E
(Scheme 4), which forms the quinoxaline 3, substituted at the