S.-M. Wang et al. / Tetrahedron Letters xxx (2015) xxx–xxx
3
products by another molecule of R3P eventually led to mixtures
of symmetric and asymmetric bis(phosphonium)ethylenes.8
HCF2CH2I and other difluoroethyl esters (like TsOCH2CF2H and
AcOCH2CF2H) were also employed to synthesize bis(phosphonium)-
ethylene salts. The negative results (such as the incomplete trans-
formation of the starting materials and the formation of massive
unknown byproducts) indicated that 2a may be the best reagent
for this reaction.
Based on the results above, we envisioned a reaction mecha-
nism in Scheme 1, which might involve SN2-type substitution,
nucleophilic addition, and b-F elimination. First, nucleophilic sub-
Figure 2. ORTEP diagrams of 4d. Ellipsoids are shown at the 50% probability level.
Hydrogens and anions are omitted for clarity.
stitution of 2a by 1 at the a-carbon gives monophosphonium salt
4. Intermediate 4 is susceptible to b-F elimination providing 5
and 6 in the presence of a second phosphine (or FÀ) because of
the favorable leaving ability of fluoride.13,16 Compound 5 would
be easily dissociated at high temperature to release 1 and HF. Inter-
mediate 6 is nucleophilically attacked by 1 to form 7, which under-
goes b-F elimination to generate 8. Reaction of 8 with excess 2a
finally affords bis(phosphonium)ethylene 3. When R1, R2, or R3 is
(para-methylphenyl)phosphine (1b) with 2a afforded 3b even at
room temperature, albeit in a low yield (entry 1, Table 2). However,
phosphines with ortho-substitution on the phenyl rings frustrated
the transformation, which gave only monophosphonium salts (4d,
4e, entries 5 and 6, Table 2).16 The structure of 4d was strictly con-
firmed by X-ray crystallography (Fig. 2).14 We speculate that the
steric hindrance of the ortho-substituents on phenyl rings might
cause the failure of the conversion.16
Using arylalkylphosphines and alkylphosphines instead of
arylphosphines as substrates, the reactions with 2a became com-
plicated (entries 13–18, Table 2). For example, treatment of 1l with
2a at 120 °C for 15 h gave monophosphonium salt 4l in 66% yield
(entry 13, Table 1),16 whereas 1m reacting with 2a at 120 °C for
8 h provided bis(phosphonium)ethylene 3m in 72% yield (entry
14, Table 2). Dimethyl(phenyl)phosphine (1n) reacted with 2a at
room temperature for 72 h, after standard workup, affording crude
4n in a moderate yield (entry 15, Table 2). Heating the reaction
mixture to 120 °C, however, gave a chaotic mixture of salts. Inter-
estingly, the reaction of tricyclohexylphosphine (1o) with 2a at
room temperature for 12 h furnished 4o in 55% yield, while the
similar mixture reacted at 120 °C for 8 h provided 3o in 63% yield
(entries 16 and 17, Table 2). In the case of P(n-Bu)3 (1p), no pure bis
(phosphonium)ethylene product was obtained under the standard
reaction conditions according to NMR spectroscopy analysis (entry
18, Table 2). The elusive electronic and steric effects that arose
from the small variations in alkyl groups made the reactions a little
difficult to be sought out.
an alkyl group, the tendency of a-deprotonation of phosphoniums
at R1, R2, or R3 site, competing with CH2CF2H group, is rationalized,
which might illustrate the frustrated transformation of alky-
larylphosphines and alkylphosphines. Nonetheless, the whole
mechanism of the reaction is still unclear.
In conclusion, we have developed a facile method to the synthe-
sis of vinylenebis(trialkylphosphonium) salts from aryl-, alkyl-,
and arylalkylphosphines by using TfOCH2CF2H as reagent. The
reactions proceeded at 120 °C under solvent-free conditions to
afford (E)-bis(phosphonium)ethylenes in good to high yields. The
reaction temperature and the substituents on the phenyl rings of
phosphines greatly influence the reaction. Phosphines bearing
para- and meta-substituents on the phenyl rings favored the for-
mation of bis(phosphonium)ethylenes, whereas the substrates
with ortho-substitution on phenyl rings gave only monophospho-
nium salts.16 The reactions of arylalkylphosphines and alkylphos-
phines are elusive but some of them can also provide the desired
products. Further mechanistic study and the application of these
vinylenebis(trialkylarylphosphonium) salts are ongoing in our lab.
Acknowledgements
In addition, extensive efforts were directed to the synthesis of
asymmetric [Ph3PCH@CHPR3]2+ꢀ2XÀ from [Ph3PCH2CF2H][OTf]
and R3P (R = p-MeC6H4, m-MeC6H4, p-MeOC6H4, Cy). The reactions
did occur, but unfortunately, no pure asymmetric vinylenebis(tri-
alkylarylphosphonium) salts were isolated. The possible nucle-
ophilic vinylic substitution of the phosphonium groups of the
Financial supports of this research by Wuhan University of
Technology, the Natural Science Foundation of Hubei Province
(China) (2015CFB176), and the ‘Chutian Scholar’ Program from
Department of Education of Hubei Province (China) are gratefully
acknowledged.
R1
P
R3
R1
P
R3
H
R2
F
5
F
R2
R1
P
R1
P
R1
P
R3
O
O
R1
P
R3
F
F
R2
F
R2
S
R2
R2
+
O
CF3
R3
H
F
R3
2a
OTf
OTf
1
6
4
R1
R2
R3
R1
F
R1
R2
R3
2 OTf ( with trace fluorides)
R1
P
R3
R1
P
R3
R1
P
R3
P
P
R2
P
2a
R2
R2
R3
F
R2
OTf
,
OTf
3
7
8
R1
P
R1
P
R3
heat
R2
+
HF
H
R2
F-
R3
5
Scheme 1. Proposed mechanism for the solvent-free synthesis of bis(phosphonium)ethylenes from phosphines (1) and 2a.