(24.0 mmol, 48.0 mM) to DCE-CF3 (24.0 mmol, 11.2 mM), giving
iron-limited reaction conditions, yielded 50 3% 2, 29 1%
3 and 15
1% 4, as determined by 31P NMR (integrated to
an internal standard, tetrabutylammonium hexafluorophosphate,
error is standard deviation, n = 3). Under conditions of excess
iron achieved by slow addition of DCE-CF3 to 1, the amount
of 2 decreased to 26 2% while 3 and 4 increased to 36 3%
and 36 2%, respectively. All three products were unreactive
with DCE-CF3. The only volatile species detected other than
unreacted DCE-CF3 under either set of conditions was trace 3,3,3-
trifluoropropyne, indicating its role as a reactive intermediate.
An insoluble and uncharacterized black solid was observed in
both cases. Based on the demonstrated propensity of 3,3,3-
trifluoropropyne to polymerize13 we believe this product to be
the corresponding polymer.
The reaction stoichiometry for DCE-(CF3)2 and DCE-CF3 was
determined by reacting each substrate with 1 at various ratios
ranging from 0 : 1 to 2 : 1 (1:substrate) and analyzing the remaining
pool of substrate. These results indicate reaction stoichiometries
of 1 : 1 with DCE-(CF3)2 and 2 : 1 with DCE-CF3 (Fig. 4).
Scheme 1 Proposed mechanism for reaction of 1 with DCE-CF3.
literature precedent for ligand exchange at L5Fe.15,16 Oxidative
C–Cl addition leads to an unobserved intermediate that can
undergo syn- (Z isomer) or anti- (E isomer) b-Cl-elimination,
to give 2 and the corresponding elimination products, 3,3,3-
trifluoropropyne or hexafluoro-2-butyne. The reaction contin-
ues in the case of DCE-CF3 with deprotonation of 3,3,3-
trifluoropropyne by a second equivalent of 1 to give 3 and
the corresponding 3,3,3-trifluoropropylide anion. The propylide
anion can then substitute for the inner-sphere chloride of 2 to give
final products 3 and 4.
These final steps of the mechanism follow the method for
independent preparation of 4 (Fig. 3) from the reaction of 2
with terminal alkynes in the presence of base. We were able to
test these proposed steps through independent reactions. Evidence
for the abstraction of the acidic proton of 3,3,3-trifluoropropyne
(pKa of 18.517) by 1 comes from observation of the characteristic
NMR hydride resonance of 312 upon reaction of 1 with 3,3,3-
trifluoropropyne. No 4 was formed in this reaction; however,
upon addition of 2 to the same reaction mixture, 4 was detected,
supporting its formation through this pathway.
The overall stoichiometry of 2 : 1 for this proposed mechanism
agrees with the data obtained from experiments (Fig. 4), with the
second equivalent of 1 required because of its fast reaction with
3,3,3-trifluoropropyne. This mechanism agrees with the observa-
tion that reaction with DCE-(CF3)2 requires only one equivalent
of 1 since the non-terminal alkyne produced, hexafluoro-2-butyne,
is not subject to deprotonation.
This mechanism also accounts for the difference in the ratios
of products in the reaction with DCE-CF3 depending on the
order of reactant addition. The product suite favors 2 when 1
is limited, while conditions where two equivalents of 1 are readily
available support the formation of 4 with the concurrent depletion
of 2. In addition, the predicted ratio of 1 : 1 for 3:4 is observed
when 1 is readily available. Deviations from this ratio in the
iron-limited case may be explained by the expected instability of
the 3,3,3-trifluoropropylide anion that may react through other
routes before reaction with 2, including polymerization. While the
C–Cl activation mechanism better accounts for the observations, it
should be noted that a combination of the deprotonation pathway
and C–Cl activation pathway could be operating.
Fig. 4 Fraction of substrate remaining following reaction versus the initial
ratio of 1 to substrate, DCE-(CF3)2 ᭢ and DCE-CF3
.
᭹
A mechanism for the dechlorination reaction of 1 with DCE-
(CF3)2 and DCE-CF3 must account for the following key obser-
vations: (1) The reaction of 1 with DCE-CF3 generates three Fe-
containing products: 2, 3 and 4; (2) the ratio of those products are
dependent on the initial reactant ratio; (3) complete consumption
of DCE-CF3 requires two equivalents of 1 and (4) reaction of 1 and
DCE-(CF3)2 has a 1 : 1 stoichiometry and gives one iron product,
2.
In the case of DCE-CF3, two reasonable mechanistic possibil-
ities are either deprotonation of the acidic DCE-CF3 hydrogen
(estimated pKa = 2214) or C–Cl oxidative addition. Direct depro-
tonation of DCE-CF3 by 1, a strong base capable of deprotonating
acidswhosepKa is £25,12 wouldgive3andtheeliminationproduct,
1-chloro-3,3,3-propyne. The resulting chloro-propyne could then
oxidatively add to a second equivalent of 1 to give 4. Although
this route is viable, it does not account for the observation of 2 or
3,3,3-trifluoropropyne.
This system promotes elimination at both DCE-CF3 and DCE-
(CF3)2 yet gives a varied product suite, depending on whether
an internal or terminal alkyne is produced. It is noteworthy that
if similar processes are occuring at bulk iron(0) surfaces, the
formation of iron-alkynyl products observed in this study suggests
We therefore suggest a C–Cl activation mechanism, depicted in
Scheme 1 with DCE-CF3, for both substrates. This mechanism is
initiated by dissociation of a phosphine arm on either the dmpm
t
or SiP3 ligand. We favor a dissociative substitution based on
This journal is
The Royal Society of Chemistry 2011
Dalton Trans., 2011, 40, 1646–1648 | 1647
©
Thoreson, Kristen A.
