We commenced our investigation on the reaction with
N,
N-dimethyl enaminone
1a and
gem-difluorocyclopropene
2a in DCE in the presence of H
2O at 110 ℃ under a N
2 atmosphere by using [Cp*RhCl
2]
2 (Cp* = pentamethylcyclopentadienyl) as a catalyst, AgSbF
6 and HOAc as additives, and Cu(OAc)
2 as an oxidant (
Table 1, entry 1). An initial attempt gave desired product
3a in 20% yield. Other catalysts commonly used in C(sp
2)—H activation reactions, such as [Cp*Rh(CH
3CN)
3](SbF
6)
2, [RuCl
2(
p-cymene)]
2, [Cp*Co(CO)I
2], Cp*Rh(OAc)
2 and [Cp*IrCl
2]
2, were screened in this catalytic system (entries 2−6). The results indicated that the former demonstrated a higher catalytic ability, enhancing the yield to 43% (entry 2); in sharp contrast, the latter four completely inhibited the reaction process, and no desired product was observed. Taking [Cp*Rh(CH
3CN)
3](SbF
6)
2 as the catalyst, other reaction parameters, including the additive, oxidant, solvent and reaction temperature, were investigated to further improve the efficiency of this transformation. Exchanging HOAc for 1-adamantanecarboxylic acid (1-AdCO
2H) or pivalic acid (HOPiv) could drive this reaction but lowered the yield of
3a (entries 7 and 8). When using CsOAc or Na
2CO
3 as the base, this reaction did not proceed (entries 9 and 10), showing that the acids as additives are beneficial for this transformation whereas the bases suppressed this reaction process. Next, different oxidants, namely AgOAc,
tert-butyl hydroperoxide (TBHP), O
2, Ag
2CO
3, and Ag
2O, were tested (entries 11–15): the former three displayed poor oxidative performances, and their use in fact dramatically suppressed the formation of
3a (entries 11–13); in contrast, and to our delight, the latter two oxidants drove this reaction to work more efficiently, delivering higher yields of
3a (entries 14 and 15), compared with Cu(OAc)
2—and of these two oxidants, Ag
2CO
3 proved to be a better choice for this domino process (79%, entry 14). After that, the yields of product
3a associated with [Cp*Rh(CH
3CN)
3](SbF
6)
2 in the presence of HOAc and Ag
2CO
3 in several other aprotic solvents were summarized as follows (entries 16–22): dichloromethane (DCM, 83%), CHCl
3 (9%), xylene (34%),
orthodichlorobenzene (ODCB, 47%), 1, 4-dioxane (51%), 2, 2, 2-trifluoroethanol (TFE, 0%), and ethyl acetate (EA, 34%), among which DCM gave a higher yield than other solvents (entry 16). Reducing the reaction temperature led to an incomplete consumption of starting materials, lowering the yield to 51% (entry 23). Without HOAc or H
2O, the yield of
3a severely declined to 29% and 57% respectively (entries 24 and 25), showing that additional water is necessary for this transformation. The reaction did not proceed in the absence of the Rh catalyst (entry 26). Moreover,
N,
N-diethyl (
1ab) and pyrrolidin-1-yl (
1ac) enaminones were chosen to examine the effect of substituents linked by nitrogen atoms on the efficiency of the transformation. The results revealed that the target compound
3a could be generated, but with decreased yields (entries 27 and 28).