themselves capable of coordination to palladium so giving the
required cis geometry.
to the catalytic solution containing palladium acetate (45 mg,
0.02 mmol), methanesulfonic acid (30 mL, 0.4 mmol) and the
phosphine in dry methanol (10 cm3) being cannulated in under
nitrogen. The autoclave was then pressurised to 60 bar using ethene
and carbon monoxide (1 : 1) and heated to 85 ◦C during 10 min.
The pressure was allowed to fall during the reaction period (batch
mode). After 2.5 h, the reaction was quenched by rapid cooling of
the autoclave and the unreacted gases vented. The reaction slurry
was then filtered and the liquid kept for GC analysis. In the case
of polymer production, the solid was washed with fresh methanol,
dried and weighed. The catalyst activity for polymer formation was
determined both by the weight of the polymer produced (g PK (g
Pd h)-1) and by average turnover frequency over 2.5 h (mol ethene
introduced (mol Pd h)-1); the mass of the polymer was divided by
56 (CO + ethene) to give an approximation of the turnover number.
This approximation is reasonable for longer chain polymers. This
number is included in an attempt to obtain a comparison with the
rates of formation of methyl propanoate. The activity for methyl
propanoate and co-oligomer formation was determined by gas
chromatography.
The most significant and relevant consideration of unidentate
phosphines, out with the conventional literature, is work patented
by Nozaki (Shell),16 where catalysts of general formula PR1R2R3
are proposed to be active for the production of polyketone, where
R1 is an alkyl group and R2, R3 are aryl groups.
We now report our investigation on unidentate phosphines in
the methoxycarbonylation of ethene, trying to rationalise our
results in terms of steric and electronic effects of the phosphines.
Experimental
All manipulations and reactions were carried out using standard
Schlenk techniques, using oven-dried glassware. All experiments
were performed under a nitrogen atmosphere, dried through a
Cr(II)/silica packed glass column. Air-sensitive liquids were stored
and handled under nitrogen. Air-sensitive solids were stored in the
glove box and handled under nitrogen.
1H, 13C and 31P NMR spectra were recorded using automated
Bruker AM 300/400 spectrometers. Broad band decoupling was
used for 31P spectra and 13C spectra, for which distortionless
enhancement by polarisation transfer (DEPT) was also used. Gas
chromatography was performed using a Hewlett-Packard 6890
series gas chromatograph, equipped with an Agilent 6890 series
injector. Both quantitative analysis by a flame ionisation detector
(GC-FID) and qualitative analysis by a Hewlett-Packard 5973 se-
ries mass selective detector (GC-MS) were performed. A Hewlett-
Packard Chemstation allowed the computerised integration of
peak areas.
CO and ethene were purchased from BOC gases. Methanol
was distilled over magnesium ethoxide under argon. Toluene
was purified using an Innovative Technologies, UK system. All
reagents were purchased from Aldrich and used as received un-
less otherwise stated. [(COD)PdMeCl],17 [Pd(PPhMe2)2MeCl],27
[Pd(PMe)2MeCl]27 and [Pd(PPh3)2MeCl]28 were prepared accord-
ing to the literature. Methanesulfonic acid was degassed before
use.
Cone angles for phosphines were taken from the literature or
calculated using a literature method.18 Electronic parameters were
based on the IR spectrum of [Ni(CO)3PR3] from the literature or
calculated.18
Results and discussion
Triaryl-, dialkylphenyl-, alkyldiphenyl- and trialkyl-phosphines
have been examined in this study. The same catalytic conditions,
pressure, temperature, reaction time, catalyst concentration as well
as the same acid promoter (MeSO3H) were applied, for ease of
comparison with previous studies. The palladium to phosphine
ratio was varied from 1 : 4 to 1 : 30, along with the nature of the
phosphine in order to focus on the steric and electronic effects of
the ligands. The results are collected in Table 1.
This related family of phosphines contains triphenylphosphine,
which has historically been widely studied, particularly by early
researchers such as Sen, who showed that methyl propanoate was
produced along with small amounts of short chain oligomer.13,19
Thus, this gives a reference point from which we can rationalise
the observed product selectivity of related triarylphosphine con-
taining catalysts. When triphenylphosphine (Pd–P = 1 : 4) is used,
both methyl propanoate and polyketone are formed, but at very
low rates (runs 1–3). This observation is entirely consistent with
Drent’s observations20 that at low phosphine to palladium ratios,
some activity to polyketone can be observed, because cis–trans
isomerisation in the acyl intermediate is slower.
For triphenylphosphine (runs 1–3), tris-(4-fluorophenyl)pho-
sphine (runs 4–6) and tris-(4-methoxyphenyl)phosphine (runs 9–
11) a general increase in the rate of formation of methyl propanoate
can be observed from the 1 : 4 to 1 : 30 ratio for each phosphine.
This observation may be also be related to the increase in the
rate of cis–trans isomerisation. Tris-(2-methoxyphenyl)phosphine
(runs 12 and 13) totally inhibits the reaction, presumably because
the high degree of steric bulk directly around the phosphorus
atom provided by the ortho substituents prevents any monomer
molecule from coordinating to palladium in the catalytic complex
or leads to complexes containing only one phosphine ligand.
No activity is observed using the very bulky and electron poor
P(C6F5)3 (runs 7 and 8).
Trans-[Pd(PPh2Me)2MeCl]
A solution of [(COD)PdMeCl] (5 mg, 0.019 mmol, COD = 1,5
cyclooctadiene) with 2 equiv. of PPh2Me (6.9 mL, 0.037 mmol)
in d8-toluene was syringed into an NMR tube and was shaken
for 10 min. 1H NMR (400.13 MHz, C7D8, Me4Si, 193 K) d/ppm:
0.35 (3H, t, JPH 6, Pd–CH3), 1.15 (6H, quin, JPH 8, P–CH3), 6.99–
7.03 (8H, br s, Ar–H), 7.15–7.20 (4H, br s, Ar–H) and 7.50–7.59
(8H, br m, Ar–H). 13C NMR (100.6 MHz, C7D8, Me4Si, 193 K)
d/ppm: 1.2 (Pd–CH3), 12.5 (P–CH3), 128.0, 129.5 and 131.9. 31
P
NMR (121.5 MHz, C7D8, Me4Si, 193 K) d/ppm: 26.6. m/z (ESI)
521.0770 (M - (Cl35). C27H29PdP2 requires 521.0779).
Catalytic runs
In addition to this steric effect, we can also observe an electronic
effect when the methyl propanoate rate is compared between the
first four phosphines in Table 1 (runs 1–11). A general increase
Catalytic reactions were carried out in identical 250 mL
HasteloyTM autoclaves. The reactor was purged with CO prior
This journal is
The Royal Society of Chemistry 2009
Dalton Trans., 2009, 872–877 | 873
©