distances of 2.179(4), 2.145(4), 2.233(4) Å],9 and Mo [2.308(5),
2.228(5), 2.453(5) Å].10
In the present complex the strong π conjugation in the allyl-
methoxy group and steric interactions between the methoxy
and the phenyl group of the diphosphine [C(31) ؒ ؒ ؒ O(1)
distance of 3.45 Å] may account for the differences observed in
the coordination Pd–allyl distances.
of Pd–diphosphine complexes containing a benzyl ligand,
synthesised on protonation of the corresponding styrene
derivatives, shows that, in all cases, the π-benzylic linkage is the
preferred coordination mode.19
On the basis of the above considerations and of the obtained
results we propose that the inactivity of P–P ligands towards
the CO/aromatic olefins copolymerisation may be due to the
inertness of the π-allylic (or π-benzylic) fragment towards CO
insertion.|| Therefore, in this case the β-hydrogen elimination
becomes the fastest reaction, yielding the compound E-1,5-
diphenylpent-1-en-3-one,2 instead of the polyketone.
The π-homoallylic structure is retained in solution, as
deduced from the 1H and 13C NMR analysis.† The main feature
1
of the H NMR spectrum of 2, measured in CD2Cl2 at room
temperature, is the lack of the two multiplets at 6.09 and 5.91
ppm typical for the olefin protons of the cyclooctene moiety.4
They are replaced by a doublet (4.83 ppm) and a multiplet (3.92
ppm), coupled to each other, in agreement with the literature,11
indicating the presence of two protons of an allylic fragment.
The signal at lower field of the two is assigned to the proton
bound to the carbon atom C1 (see Scheme 1 for the numbering
system of η3-C8H12OMe), as deduced by the clear NOE with
the methoxy group, thus indicating that the syn configuration is
also present in solution. In the 13C NMR spectrum, the signals
of carbon atoms bound to palladium (C1 and C2 at 85.6
and 69.2 ppm, respectively) have chemical shifts typical for an
allylic system.12 The signal of the third carbon atom (C8) of
the allylic moiety is at 155.0 ppm, due to the bond with the
oxygen atom of the methoxy group. This downfield shift has
already been reported for a palladium–allylic system bearing
such a substituent.10,13 These three signals also show a different
fine structure, thus confirming their different relative position
with respect to the phosphorus atoms. Actually, both the reson-
ances of C2 and C8 are a double doublet (even though that due
to C8 is not very well resolved), indicating coupling with the cis
and trans phosphorus atoms, respectively. On the other hand,
the signal of C1 is a triplet suggesting a pseudoequivalence of
the two phosphorus atoms with respect to it.
Acknowledgements
This work was supported by Ministero dell’Università e della
Ricerca Scientifica (MURST-Rome) grant no. 9903153427.
Notes and references
† Complex 2 was synthesised as previously reported,4 by adding dppp
instead of bpy. The solid obtained is yellow; yield 0.54 g, 86%. Found:
C, 53.0; H, 5.02%. C36H41OF6P3Pd requires: C, 53.84; H, 5.15%. δH (400
MHz; solvent CD2Cl2, referenced to the solvent peak versus TMS at δ
5.30; T = 293 K): 7.53–7.40 [20 H, m, Ph], 4.83 [1 H, d, H1], 3.92 [1 H,
m, H2], 3.13 [3 H, s, OCH3], 2.66 [4 H, br m, CH2-P], 2.39 [1 H, br m,
HA7], 2.02 [1 H, br m, CH2A-CH2-P], 1.88 [ 1 H, br m, HB7], 1.65 [1 H,
br m, CH2B-CH2-P], 1.45–1.31 [8 H, br m, HA,B6, HA,B5, HA,B4, HA,B3];
δC (100.5 MHz; solvent CD2Cl2, referenced to the solvent peak versus
TMS at δ 53.8; T = 293 K): 155.0 [dd, 2J(C–P) = 3.7 Hz, C8], 134.0–
129.0 [m, Ph], 85.6 [t, 2J(C–P) = 5.5 Hz, C1], 69.2 [dd, 2J(C–P) = 5.5 Hz,
2J(C–P) = 31.2 Hz, C2], 56.5 [s, OCH3], 37.5 [s, C7], 30.0 [d, C6 or C5 or
C4 or C3], 28.0–27.0 [m, CH2-P and C6 or C5 or C4 or C3], 26.0 [m, C6
or C5 or C4 or C3], 22.8 [s, C6 or C5 or C4 or C3], 18.5 [s, CH2-CH2-P];
δP (161.86 MHz; solvent CD2Cl2; H3PO4 as external reference; T = 293
K): 11.1 [d, 2J(PA-PB) = 56.5 Hz, PA or PB], 6.1 [d, 2J(PA-PB) = 56.5 Hz,
PA or PB].
‡ The ligand dppp (0.28 g, 0.68 mmol) was added to a suspension of
[Pd(η1,η2-C8H12OMe)(bpy)][PF6] (0.31 g, 0.57 mmol) in methanol (20
mL), yielding a yellow solution. After a few minutes the product
precipitated as a yellow solid. It was stirred at room temperature for 30
min, filtered off, washed with diethyl ether and dried under vacuum.
Yield: 0.27 g, 60%. Found: C, 53.0; H, 4.89%. C36H41OF6P3Pd requires:
C, 53.84; H, 5.15%.
§ Crystallography. C36H41F6OP3Pd, M = 803.00, monoclinic, space
group P21/n, a = 10.716(2), b = 18.008(2), c = 18.978(4) Å, β = 98.01(1) Њ,
V = 3626.5(11) Å3, Z = 4, Dcalc = 1.471 g cmϪ3, µ(Mo-Kα) = 0.703
mmϪ1, T = 293 K, R1 = 0.0490, wR2 = 0.1256 for 8213 unique reflec-
tions. All the calculations were performed using the WinGX System,
Ver 1.63.5 CCDC reference number 186/2140. See http://www.rsc.org/
suppdata/dt/b0/b005492p/ for crystallographic files in .cif format.
¶ When the complex 1 was used as precatalyst for the CO/ethylene
copolymerisation under the same reaction conditions reported for the
CO/styrene one,4 a productivity of 19 g CP/g Pd was obtained.
|| When complex 1 (30 mM, CD2Cl2, room temperature) was treated
with carbon monoxide (1 atm) all of the signals in the 1H NMR
spectrum broadened and a red solid precipitated. On the other hand,
when carbon monoxide was bubbled into a solution of 2 (30 mM,
1
In the H NMR spectrum, at room temperature, the signals
of the propylenic bridge of dppp and those of the protons from
H3 to H7 are broad. Decreasing the temperature (from 293 K
to 213 K) resulted only in a further broadening, while the
signals of the allylic moiety were not significantly affected by
temperature, thus indicating that they are not involved in a
dynamic process. The stability of the syn configuration seems to
be a feature of π-allylic systems substituted with a methoxy
group.10
The compound we obtained is the result of a rearrangement
of the organometallic fragment from a σ,π-enyl cooordination
mode to a π-allylic one, involving the C8-atom. Similar cyclic
palladium–allylic systems are reported in the literature, but
in no case is the C-atom bearing the alkoxy group part of
the allylic fragment.14 Therefore, the above rearrangement
implies a migration of the C–C double bond. As a first step of
the mechanism a β-hydrogen elimination from the carbon atom
substituted with the methoxy group of the C8H12OMe ligand,
either on the starting dimer or on 1, is proposed, in agree-
ment with the literature.15 Subsequent hydride-migrations/
elimination reactions may occur yielding the final product. A
similar mechanism was reported for the η4-diene→η3-allyl
rearrangement of the cycloocta-1,5-diene ligand in ruthenium–
phosphine complexes.16 Our results indicate that the σ,π→η3
rearrangement is induced by the diphosphine ligand.
1
CD2Cl2, room temperature) no change in the H NMR spectrum was
observed.
1 (a) A. Sen and T. W. Lai, J. Am. Chem. Soc., 1982, 104, 3520; (b)
T. W. Lai and A. Sen, J. Am. Chem. Soc., 1984, 3, 866; (c) E. Drent
and P. H. M. Budzelaar, Chem. Rev., 1996, 96, 663; (d) E. Drent,
J. A. M. van Broekhoven and M. J. Doyle, J. Organomet. Chem.,
1991, 417, 235.
2 M. Barsacchi, G. Consiglio, L. Medici, G. Petrucci and U. W. Suter,
Angew. Chem., Int. Ed. Engl., 1991, 30, 989.
3 G. Pietropaolo, F. Cusmano, E. Rotondo and A. Spadaro, J. Organo-
met. Chem., 1978, 155, 122.
4 A. Macchioni, G. Bellachioma, G. Cardaci, M. Travaglia, C.
Zuccaccia, B. Milani, G. Corso, E. Zangrando, G. Mestroni,
C. Carfagna and M. Formica, Organometallics, 1999, 18, 3061.
5 WinGX–A Windows Program for Crystal Structure Analysis, L. J.
Farrugia, J. Appl. Cryst., 1999, 32, 837.
When complex 2 was tested as a catalyst precursor for the
CO/ethylene and CO/styrene copolymerisations, under the
reaction conditions reported for complex 1, no formation of
polyketone was observed.¶
From a critical review of the literature, the following obser-
vations are summarised: 1. The diphosphine ligands (P–P)
promote the CO/aliphatic olefins copolymerisation, but do not
promote the CO/aromatic ones;2 2. The nitrogen-donor ligands
(N–N) promote copolymerisation of CO with both aliphatic
and aromatic olefins;1,17 3. The CO/styrene copolymerisation
promoted by N–N ligands involves an equilibrium between a
chelate and a π-benzylic intermediate;18 4. It is suggested that
in the initiation step of the CO/styrene copolymerisation the
Pd–styryl intermediate is so strongly stabilised by the π-benzylic
coordination so as to prevent CO insertion;1c 5. A series
6 D. H. Farrar and N. C. Payne, J. Am. Chem. Soc., 1985, 107,
2054.
7 A. Togni, U. Burckhardt, V. Gramlich, P. S. Pregosin and R. Salz-
mann, J. Am. Chem. Soc., 1996, 118, 1031.
3056
J. Chem. Soc., Dalton Trans., 2000, 3055–3057