Chemistry Reference
In-Depth Information
Unfortunately the conclusions put forward by
Tundo's group could not fully explain or predict
the inconsistent role of phase-transfer catalyst in
hydrodehalogenation reactions catalysed by Pd/C.
No practical guidelines could be prompted for a
rational selection of a phase-transfer catalyst for a
given task in these systems. Regioselective hydrode-
bromination was claimed by Sabahi [390], who con-
verted 1,6-dibromo-2-methoxynaphthalene into 2-
bromo-6-methoxynaphthalene using hydrogen with
a tungsten carbide/phase-transfer catalyst system.
As early as 1978, Bamfield & Quan [391] reported
a different reaction path for the hydrogenolysis of
aryl halides catalysed by Pd/C upon the addition of
various cationic surface-active agents. Using sodium
formate as the reducing agent (Scheme 10.49) it was
found that in the presence of CTAB an Ullmann-type
reaction was the major process.
Quaternary ammonium salts likewise were
endorsed for the homogeneous Pd(OAc)
2
-catalysed
homocoupling of aryl halides (Scheme 10.50). This
system, developed by Lemaire
et al
. [392], has proved
to be compatible with various sensitive functional
groups. Biheterocycles such as bipyridines also were
accessible using this methodology (Scheme 10.51).
A major shortcoming of this system was that the
palladium catalyst was reduced in the course of the
reaction and could not be recycled.
We have observed recently that one of the key
parameters controlling the selectivity of the reduc-
tion reaction of halobenzenes to homocoupling or
hydrogenolysis was the reaction temperature. Thus,
the activation energy of the Pd/C-catalysed homo-
coupling of chlorobenzene to biphenyl was mea-
sured to be 63 KJ mol
-1
, whereas
E
a
for the dehy-
droalogenation of chlorobenzene to benzene under
the same conditions was found to be an order of
magnitude smaller (6 KJ mol
-1
) [393]. These results
characterise the hydrogenolysis reaction as a mass-
transfer-controlled process and suggest that higher
temperature would favour the coupling reac-
tion. This was indeed the case: the selectivity of the
homocoupling of chlorobenzene to biphenyl was
increased from 45% at 90°C to 70% at 100°C and
85% at 110-120°C in the system containing
Pd/C/CTAB/HCOONa/NaOH [394].
We also found that the reductive coupling reaction
can be induced favourably with hydrogen gas as a
reducing agent. The hydrogen can be supplied exter-
nally or generated in situ, for instance by reaction
of zinc with water [395] or by decomposition of
formate salts in water [396]. Both hydrogen-
producing reactions were catalysed by Pd/C (Scheme
10.52).
Numerous phase-transfer agents were shown to
improve the conversion, particularly the selectivity
of the homocoupling reaction. Highest chemoselec-
tivity was monitored with CTAB, TBAB and PEG-
400. Crown ethers also were very effective [397].
Other parameters that increased the homocoup-
ling selectivity were higher Pd/C loading, higher
base amount and higher hydrogen pressure (or
hydrogen donor concentration). It was demon-
strated clearly that the Pd/C catalyst could be re-
cycled in ten consecutive runs without any loss in
activity.
We have shown evidence that the coupling of aryl
halides was the result of one electron transfer from
Pd(0) to the substrates to generate a radical anion
that ejects a chloride anion with the formation of
an aryl free radical that couples with a second aryl
radical to yield the biaryl product (Fig. 10.8).
The oxidised palladium then is reduced by hydro-
gen to regenerate Pd(0) (Scheme 10.53).
Pd/C, CTAB
HCOONa, NaOH
Cl
X
X
X
Scheme 10.49
O
O
O
Pd(OAc)
2
, Et
3
N
DMF, 115
°
C, 94 h
Pd(OAc)
2
, K
2
CO
3
DMF, IPA, 115
Br
Br
°
C, 45 h
N
NN
53% isolated yield
97% conversion
DMF = dimethylformanide
100% conversion
IPA = isopropanol
92% isolated yield
Scheme 10.50
Scheme 10.51
