Angewandte
Chemie
As mentioned above, the research groups of Brookhart
and Tilley independently reported the catalytic activity of two
rhodium complexes for the dehydrocoupling of phosphanes
to the corresponding diphosphanes. The catalytic activity of
complex 3 for such reactions was tested using PHPh2 as
a model substrate in toluene at 808C with 5 mol% of the
catalyst. After 13 h, 51% conversion was observed, while
hydrogen was observed by 1H NMR spectroscopy. A plausible
catalytic cycle for this reaction is shown in Scheme 3. Hydro-
gen could be eliminated from complex 3 by addition of PHPh2
to give the transient species A, which contains two cis-
phosphanide ligands. Reductive elimination of the diphos-
phane from A upon reaction with a new molecule of
phosphane would regenerate the catalyst. Nonetheless,
other possibilities involving s-bond metathesis steps cannot
be excluded.
Although a hydrogen acceptor is not strictly necessary, the
reaction is considerably faster under an ethylene atmosphere
(6 bar). Under these conditions, a 100% conversion was
achieved within 5 h at 808C (Figure 3).[24] Thus, the catalytic
activity of 3 under non-optimized conditions is better than
that observed for other rhodium compounds.[9] This reaction
could be rationalized if compound 3 engaged in competitive
alkene insertion to give the ethyl intermediate B. Then, the
ethyl ligand could be protonated by the incoming phosphane
to give A and ethane, thus providing an easier way for the
catalysis to proceed (Scheme 3). Indeed, the formation of one
mole of ethane per mole of the diphosphane was simulta-
neously observed and complex 3 was the sole rhodium
complex detected during and at the end of the catalysis.
It is remarkable that, as the reaction proceeds, PEtPh2
Figure 3. Conversion [%] versus time [h] for the dehydrocoupling of
PHPh2 catalyzed by 3.
phosphane favors the protonation of the ethyl group in B and
ꢀ
inhibits the P C reductive elimination to PEtPh2, which
becomes operative at a low phosphane concentration.
If equimolecular mixtures (prepared in situ) of 2 and
other phosphanes such as PHPh2, PMePh2, and PMe2Ph are
used as catalyst precursors, the catalysis proceeds with
identical results in all cases. Under 6 bar of ethylene, a full
conversion was observed in 7 h and the products were found
ꢀ
to be Ph2P PPh2 (80%) and PEtPh2 (20%). Inspection of the
catalyst precursors by NMR spectroscopy revealed the
quantitative formation of hydride phosphanide complexes
[Rh(Tp)(H)(L)(PPh2)] (L = PHPh2, 6; PMe2Ph, 7; PMePh2,
8) similar to 3 (see the Supporting Information). After
addition of the substrate (PHPh2), pressurizing with ethylene,
and warming at 808C, the sole rhodium species present was
[Rh(Tp)(H)(PHPh2)(PPh2)] (6), which clearly shows that the
catalysis was performed with 6.
The catalytic cycle with complex 6 fits with that proposed
for 3 (Scheme 3, L = PHPh2). An increase in the ethylene
hydrophosphination product reduces the selectivity for the
dehydrocoupling reaction relative to 3. This noticeable
difference has to be attributed to a more difficult protonation
of the ethyl group in B with L = PHPh2 than for L = PMe3,
which favors the reductive elimination of PEtPh2. Accord-
ꢀ
appears as a product when the conversion into Ph2P PPh2 is
ꢀ
over 80%, so that a mixture of Ph2P PPh2 (95%) and PEtPh2
(5%) results at the end of the catalysis. Certainly, complex 3 is
ꢀ
ꢀ
a precatalyst for both P P and P C bond formation, although
the latter reaction is not desirable in this case. However, it is
significant that the hydrophosphination of ethylene occurs,
since unactivated olefins are essentially absent in this type of
ꢀ
ingly, the P C bond formation in B would give [Rh(Tp)(L)-
(PEtPh2)] (C), in which the secondary phosphane replaces
PEtPh2 to close the catalytic cycle. On the other hand, the
phosphanido-bridged complex [(Tp)(H)Rh(m-PPh2)Rh-
(PHPh2)2] (9) was the sole rhodium compound found at the
end of the catalysis with 6. Complex 9 was found to be inactive
for the dehydrocoupling of PHPh2, which also supports the
prominent role of mononuclear phosphanide complexes in
this reaction.
reactions.[1a,13,25] Clearly, P P and P C bond-forming reac-
tions compete in such a way that a high concentration of
ꢀ
ꢀ
Complex 9 results from the weak thermal stability of 6,
which was independently found to decompose into 9 and the
protonated species HTp, a reaction that cannot be avoided by
adding an external base such as Et3N (the syntheses and X-ray
structures of 9 and [(Tp)(H)Rh(m-PPh2)Rh(PHPh2)(PMe3)]
(10) can be found in the Supporting Information[17]).
Primary phosphanes such as PH2Ph were found to be too
reactive to allow the isolation of mononuclear complexes.
Reaction of 1 with PH2Ph in C6D6 results immediately in
a mixture containing mainly the trans diastereoisomers of
[{Rh(H)(m-PHPh)(Tp)}2], as deduced from NMR spectros-
copy. Under the catalytic conditions mentioned above, the use
of either 1 or 3 as catalyst precursors and PH2Ph as substrate
Scheme 3. Plausible catalytic cycle for the dehydrocoupling of phos-
phane and hydrophosphination of ethylene mediated by rhodium
complexes. [Rh]=Rh(Tp).
ꢀ
led to only a 1% of conversion to PhHP PHPh.
Angew. Chem. Int. Ed. 2014, 53, 1 – 5
ꢀ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3
These are not the final page numbers!
Geer, Ana M.
