Palladium complexes of the heterodiphosphine o- C6H 4(CH2PtBu2)(CH2PPh 2) are highly selective and robust catalysts for the hydromethoxycarbonylation of ethene

Article abstract of DOI:10.1021/om100049n

The coordination chemistry and ethene hydromethoxycarbonylation catalysis with the diphosphine o-C₆H₄(CH₂P ^tBu₂)(CH₂PPh₂) (L₃) is reported and the results compared with the analogous chemistry of the symmetrical diphosphines o-C₆H₄(CH₂P ^tBu₂)₂ (L₁) and o-C ₆H₄(CH₂PPh₂)₂ (L ₂). Palladium-catalyzed ethene hydromethoxycarbonylation studies under the commercial catalytic conditions are reported. The results obtained using L_1-3 as supporting ligands show that the catalysts derived from L₃ and L₁ have similar activity and selectivity for methyl propanoate (MeP). In addition, the Pd-L₃ catalyst has much greater longevity than the Pd-L₁ catalyst. Treatment of the appropriate [Pt(X)(Y)(cod)] with L₃ gave [PtCl₂(L ₃)] (3), [Pt(CH₃)₂(L₃)] (6), and [PtCl(CH₃)(L₃)] (9). At equilibrium, complex 9 is a 90:1 mixture of geometric isomers 9a (with CH₃ trans to the ^tBu₂P) and 9b (with Cl trans to the ^tBu ₂P). The fluxionality of complex 3, detected by ¹H NMR, is interpreted in terms of the conformation of the seven-membered chelate. The complexes [Pt(CH₃)(PMe₃)(L₃)]Cl (10b) and [PtH(PPh₃)(L₃)]Cl (12b) are formed as essentially single isomers with CH₃/H trans to the Ph₂P group. The palladium complexes [PdCl₂(L₃)] (13), [PdCl(CH₃)(L ₃)] (14a/14b), and [PdH(PCy₃)(L₃)]BF ₄ (15b) have been made by similar methods to their platinum analogues. The factors controlling the relative isomer stabilities are explored experimentally and computationally. The complexes [PtCl₂(L ₄)] (16) and [PtCl(CH₃)(L₄)] (17a/17b) where L₄ = o-C₆H₄(CH₂PⁿBu ₂)(CH₂PPh₂) are reported, and the geometric isomers of 17 are almost isoenergetic. The crystal structures of 3, 14a, 15b, and 16 have been determined by X-ray crystallography. DFT calculations on complexes of the type [Pt(X)(Y)(L₃)] gave only small calculated differences in energy between the geometrical isomers (0-4 kcal mol ^-1), which are consistent with the experimental observations. It is suggested that repulsive intramolecular HAAAH interactions (between the Pt-CH₃ and PC(CH₃)₃ groups) determine which isomer predominates. The reasons for the favorable catalytic properties of the Pd-L₃ catalyst are probed by ¹³CO reactions with the model complexes 9a/9b and 14a/14b, and the structures of the resulting acyl complexes are assigned on the basis of ¹³C and ³¹P NMR and IR spectroscopy. From these studies, it is suggested that the reason for the Pd-L₃ catalyst resembling the Pd-L₁ catalyst in terms of selectivity is that the crucial acyl intermediates are similar.

Full text of DOI:10.1021/om100049n

2292 Organometallics 2010, 29, 2292–2305
DOI: 10.1021/om100049n
Palladium Complexes of the Heterodiphosphine
o-C6H4(CH2PtBu2)(CH2PPh2) Are Highly Selective and Robust
Catalysts for the Hydromethoxycarbonylation of Ethene
Tamara Fanjul,† Graham Eastham,‡ Natalie Fey,† Alex Hamilton,†
A. Guy Orpen,† Paul G. Pringle,*,† and Mark Waugh‡
†School of Chemistry, University of Bristol, Cantocks Close, Bristol BS8 1TS, U.K., and
‡Lucite International, Lucite International Technology Centre, PO Box 90, Wilton, Middlesbrough,
Cleveland, TS6 8JE, U.K.
Received January 21, 2010
The coordination chemistry and ethene hydromethoxycarbonylation catalysis with the diphos-
phine o-C6H4(CH2PtBu2)(CH2PPh2) (L3) is reported and the results compared with the analogous
chemistry of the symmetrical diphosphines o-C6H4(CH2PtBu2)2 (L1) and o-C6H4(CH2PPh2)2 (L2).
Palladium-catalyzed ethene hydromethoxycarbonylation studies under the commercial catalytic
conditions are reported. The results obtained using L1-3 as supporting ligands show that the catalysts
derived from L3 and L1 have similar activity and selectivity for methyl propanoate (MeP). In
addition, the Pd-L3 catalyst has much greater longevity than the Pd-L1 catalyst. Treatment of the
appropriate [Pt(X)(Y)(cod)] with L3 gave [PtCl2(L3)] (3), [Pt(CH3)2(L3)] (6), and [PtCl(CH3)(L3)] (9).
At equilibrium, complex 9 is a 90:1 mixture of geometric isomers 9a (with CH3 trans to the tBu2P) and
9b (with Cl trans to the tBu2P). The fluxionality of complex 3, detected by 1H NMR, is interpreted in
terms of the conformation of the seven-membered chelate. The complexes [Pt(CH3)(PMe3)(L3)]Cl
(10b) and [PtH(PPh3)(L3)]Cl (12b) are formed as essentially single isomers with CH3/H trans to the
Ph2P group. The palladium complexes [PdCl2(L3)] (13), [PdCl(CH3)(L3)] (14a/14b), and [PdH(PCy3)-
(L3)]BF4 (15b) have been made by similar methods to their platinum analogues. The factors
controlling the relative isomer stabilities are explored experimentally and computationally. The
complexes [PtCl2(L4)] (16) and [PtCl(CH3)(L4)] (17a/17b) where L4 = o-C6H4(CH2PnBu2)(CH2PPh2)
are reported, and the geometric isomers of 17 are almost isoenergetic. The crystal structures of 3, 14a, 15b,
and 16 have been determined by X-ray crystallography. DFT calculations on complexes of the type
[Pt(X)(Y)(L3)] gave only small calculated differences in energy between the geometrical isomers (0-4 kcal
mol-1), which are consistent with the experimental observations. It is suggested that repulsive intra-
molecular H H interactions (between the Pt-CH3 and PC(CH3)3 groups) determine which isomer
3 3 3
predominates. The reasons for the favorable catalytic properties of the Pd-L3 catalyst are probed by 13CO
reactions with the model complexes 9a/9b and 14a/14b, and the structures of the resulting acyl complexes
are assigned on the basis of 13Cand31P NMR and IR spectroscopy. From these studies, it is suggested that
the reason for the Pd-L3 catalyst resembling the Pd-L1 catalyst in terms of selectivity is that the crucial
acyl intermediates are similar.
Introduction
For example, the highly active catalyst derived from bis(di-
tert-butylphosphino)xylene (L1) produces MeP in 99.9%
selectivity,2 and in 2008, Lucite commercialized a process
founded on this chemistry. By contrast, the catalyst derived
from the diphenylphosphino analogue L2 has low activity
and gives PK in ca. 90% selectivity (vide infra).
The carbonylation of ethene (Scheme 1) in the presence of
methanol is an excellent example of an atom-efficient homo-
geneous catalysis that has been commercialized for the
production of two commodity chemicals: methyl propanoate
(MeP) and polyketone (PK).1 The best catalysts for ethene
carbonylations are palladium(II)-diphosphine complexes
where the selectivity for MeP or PK is controlled by the
ligand backbone and the substituents on the P atoms.2-7
*Corresponding author. E-mail: paul.pringle@bris.ac.uk.
(1) Drent, E.; Buzelaar, P. H. M. Chem. Rev. 1996, 96, 663, and
references therein.
(2) Clegg, W.; Eastham, G. R.; Elsegood, M. R. J.; Tooze, R. P.;
Wang, X. L.; Whiston, K. Chem. Commun. 1999, 1877.
Naıvely, it might be predicted that catalysts derived from
¨
the ditopic ligand L3 would give mixtures of MeP, oligomers,
and PK. However, we show here that the L3-Pd catalyst
r
pubs.acs.org/Organometallics
Published on Web 04/23/2010
2010 American Chemical Society

Article
Organometallics, Vol. 29, No. 10, 2010 2293
Scheme 1
Scheme 2
Chart 1
The results provide insights into the role of stereoelectronic
effects in unsymmetrical chelate complexes that may have
important consequences for the design of bidentate ligands
for catalysis.
gives MeP as selectively as the commercial Pd-L1 catalyst
and that the Pd-L3 catalyst is significantly more stable than
the Pd-L1 catalyst.8 In order to rationalize the excellent
catalytic results obtained with Pd-L3 complexes, it was
considered important to understand the stereoelectronic
factors governing the relative stabilities of the geometric
isomers that would be involved in every putative intermedi-
ate in the mechanism. We report here a combined spectro-
scopic, structural, and computational investigation of
isomeric model complexes of the type denoted a and b in
Chart 1 where M = Pt or Pd and R = H, CH3, or COCH3.
Results and Discussion
Ligand Synthesis. The synthetic route to the unsymmetri-
cal diphosphine L3 (Scheme 2) is a modification of literature
procedures for the synthesis of unsymmetrical o-xylenyl
diphosphines.9 Ligand L3 is a white crystalline solid that
has been fully characterized (see Experimental Section).10
The 31P NMR spectrum of L3 showed two singlets at -14.9
(PPh2) and 26.3 (PtBu2) ppm, and so, not surprisingly, 5JPP is
less than the line width of ca. 4 Hz.
The two P-donors in ligand L3 would be expected to have
considerably different coordination properties: RPtBu2 are
more bulky and better σ-donors than the corresponding
RPPh2.11 It has been shown12 that the JPSe for R3PdSe is
a reliable measure of the donor properties of R3P. The di-
selenides of L1-3 were generated by treatment of the diphos-
phines with KSeCN in MeOH (see Experimental Section),
and the values of 1JPSe (in CDCl3) for L3(Se)2 of 694 and 724
Hz are the same as those determined for the corresponding
symmetrical L1(Se)2 and L2(Se)2, showing that, in nonche-
lated L3, the donors do not appear to influence each other
significantly.
Hydromethoxycarbonylation Catalysis. Ethene carbonyla-
tion was carried out with palladium catalysts derived from
L1-3 under the same conditions (see Experimental Section),
and the results are given in Table 1. As expected from
previous reports,2 the catalyst derived from ligand L1 pro-
duces MeP at high rate and with very high selectivity. Under
the conditions used here, the catalyst derived from ligand L2
had very low activity and produced a mixture of solid
polyketone, liquid oligomers, and traces of MeP. Strikingly,
when ligand L3 was employed, MeP was formed in >99.5%
(3) (a) Eastham, G. R.; Tooze, R. P.; Wang, X. L.; Whiston K. (ICI)
World Patent WO 96/19434, 1996. (b) Drent, E.; Kragtwijk, E. (Shell) Eur.
Patent EP495,548, 1992. (c) Pugh, R. I.; Drent, E.; Pringle, P. G. Chem.
Commun. 2001, 1476. (d) Drent, E.; Kragtwijk, E.; Pello, D. H. L. (Shell)
Eur. Patent EP495,547, 1992. (e) Drent, E.; Pello, D. H. L.; Suykerbuyk,
J. C. L. J.; van Gogh, J. B. (Shell) World Patent WO5354, 1994. (f) Doherty,
S.; Robins, E. G.; Knight, J. G.; Newman, C. T.; Rhodes, B.; Champkin, P. A.;
Clegg, W. J. Organomet. Chem. 2001, 640, 182. (g) Knight, J. G.; Doherty,
S.; Harriman, A.; Robins, E. G.; Betham, M.; Eastham, G. R.; Tooze, R. P.;
Elsegood, M. R. J.; Champkin, P.; Clegg, W. Organometallics 2000, 19,
4957. (h) Gusev, O. V.; Kalsin, A. M.; Peterleitner, M. G.; Petrovskii, P. V.;
Lyssenko, K. A.; Akhmedov, N. G.; Bianchini, C.; Meli, A.; Oberhauser, W.
Organometallics 2002, 21, 3637. (i) Bianchini, C.; Meli, A.; Oberhauser, W.;
van Leeuwen, P. W. N. M.; Zuideveld, M. A.; Freixa, Z.; Kamer, P. C. J.;
Spek, A. L.; Gusev, O. V.; Kalsin, A. M. Organometallics 2003, 22, 2409. (j)
Gusev, O. V.; Kalsin, A. M.; Petrovskii, P. V.; Lyssenko, K. A.; Oprunenko, Y.
F.; Bianchini, C.; Meli, A.; Oberhauser, W. Organometallics 2003, 22, 913.
(k) Vautravers, N. R.; Cole-Hamilton, D. J. Dalton Trans. 2009, 2130.
(4) Drent, E.; van Broekhoven, J. A. M.; Doyle, M. J. J. Organomet.
Chem. 1991, 417, 235.
(5) (a) Dossett, S. J.; Gillon, A.; Orpen, A. G.; Fleming, J. S.; Pringle,
P. G.; Wass, D. F.; Jones, M. D. Chem. Commun. 2001, 699. (b)
Bianchini, C.; Lee, H. M.; Meli, A.; Moneti, S.; Vizza, F.; Fontani, M.;
Zanello, P. Macromolecules 1999, 32, 4183. (c) Bianchini, C.; Lee, H. M.;
Barbaro, P.; Meli, A.; Moneti, S.; Vizza, F. New J. Chem. 1999, 23, 929. (d)
€
Lindner, E.; Schmid, M.; Wald, J.; Queisser, J. A.; Geprags, M.; Wegner, P.;
Nachtigal, C. J. Organomet. Chem. 2000, 602, 173. (e) Jiang, Z.; Sen, A.
Macromolecules 1996, 27, 7215. (f) Verspui, G.; Papadogianakis, G.;
Sheldon, R. A. Chem. Commun. 1998, 401.
(6) (a) Nozaki, K. US Patent 3689460, 1972. (b) Nozaki, K. US Patent
3694412, 1972. (c) Sen, A.; Lai, T. J. Am. Chem. Soc. 1982, 104, 3520. (d)
Sen, A.; Lai, T. Organometallics 1984, 3, 866. (e) Drent, E. US Patent,
4835250, 1989. (f) Keim, W.; Mass, H.; Mecking, S. Z. Naturforsch., B:
Chem. Sci. 1995, 50b, 430. (g) Smith, G.; Vautravers, N. R.; Cole-Hamilton,
D. J. Dalton Trans. 2009, 872.
(7) van Leeuwen, P. W. N. M.; Zuideveld, M. A.; Swennenhuis, B. H.
G.; Freixa, Z.; Kamer, P. C. J.; Goubitz, K.; Fraanje, J.; Lutz, M.; Spek,
A. L. J. Am. Chem. Soc. 2003, 125, 5523.
(9) (a) Leone, A.; Consiglio, G. Helv. Chim. Acta 2005, 88, 210.
(b) Rucklidge, A. J.; Morris, G. E.; Slawin, A. M. Z.; Cole-Hamilton, D. J.
Helv. Chim. Acta 2006, 89, 1783.
(10) Ligand L3 appeared in a table of data on the hydromethoxycar-
bonylation of styrene. It was reported to give a catalyst of low activity
and selectivity. The synthesis and characterization of L3 were not
reported. See: Ooka, H.; Inoue, T.; Itsuna, S.; Masato, M. Chem.
Commun. 2005, 1173.
(11) (a) Tolman, C. A. Chem. Rev. 1977, 77, 313. (b) Gusev, D. G.
Organometallics 2009, 28, 763.
(8) Eastham, G. R.; Waugh, M.; Pringle, P. G.; Fanjul Solares, T.
World Patent WO2010001174, 2010.
(12) Muller, A.; Otto, S.; Roodt, A. Dalton Trans. 2008, 650.

2294 Organometallics, Vol. 29, No. 10, 2010
Fanjul et al.
Table 1. Catalyst Selectivity and Productivitya
Scheme 3
ligand
MeP
oligomers
PK
TON/103
b
L1
L2
L3
>99.9%
10%
>99.5%
<0.1%
22%
<0.5%
43.9
0.2
30.7
68%
b
a See Experimental Section for details of the conditions. b Not
detected.
Table 2. Catalyst Lifetime Studies for Hydromethoxycarbonylationa
weight
gain/g
cumulative
TON/103
ligand
run
L1
1
2
3
1
2
3
4
5
6
7
8
9
10
246.1
219.3
11.9
171.7
187.5
158.7
128.9
95.9
85.3
59.0
43.6
28.1
44.0
83.2
85.3
30.7
64.2
92.5
115.5
132.6
147.8
158.3
166.1
171.1
175.4
L3
24.0
Chart 2
a Clear exit solutions were observed in all cases with no evidence of
polyketone formation.
selectivity and the activity was comparable with that ob-
tained using L1.
Catalyst lifetime studies with ligands L1 and L3 were
carried out as follows, and the results are given in Table 2
At the end of each catalytic run, the volatiles were removed,
MeOH was added, and this solution was then put back into
the autoclave and retested under the same conditions. The
catalyst derived from ligand L3 showed superior retention of
catalyst life compared to L1: after 3 runs, the catalyst
maintained over 90% of its initial activity with L3 but just
5% with L1. Even after 10 consecutive runs, over 14% of
initial catalytic activity was retained using L3 (Table 2).
A simplified hydride mechanism for ethene hydromethoxy-
carbonylation catalyzed by diphosphine-palladium complexes
is shown in Scheme 3, although the individual steps in the
mechanism are still to be fully defined.13 The bulk and large bite
angle of ligand L1 will promote the migratory insertion steps i
and iii.14 The product is determined in step iv where either (a)
MeP is formed via intermolecular or intramolecular attack by
MeOH on the coordinated acyl7,15 or (b) the MeOH is displaced
by ethene followed by acyl migration, leading to CO/C2H4
co-oligomers or PK.
adjacent to the PtBu2 group that, in step iv, inhibits access of
C2H4 at the expense of the less bulky MeOH.14 Conversely,
the explanation for the PK selectivity of catalysts derived
from ligand L2 is associated with the lower steric demands of
the PPh2 groups, facilitating C2H4 coordination at step iv. In
summary, the rate and chemoselectivity for this reaction is
controlled mainly by steric factors: bulky ligands with large
bite angles promote CO insertion into Pd-alkyl bonds and
termination by methanolysis.14-18
When the P-donors of the bidentate are not the same (i.e.,
P ¼ P0 in Scheme 3), as with ligand L3, two isomers are
possible for each intermediate in the cycle and at the migra-
tory insertion steps i and iii the metal configurations are
inverted. The high MeP selectivity of the catalyst derived
from L3 is consistent with isomer A1 (with the PtBu2 group cis
to the MeOH) prevailing over isomer A2 (Chart 2). It is
possible that isomers A1 and A2 are both present but that A1
predominates or reacts more rapidly.
The high selectivity of catalysts derived from bulky dipho-
sphine L1 for MeP has been ascribed to the restricted space
The intermediates A1 and A2 contain one strongly σ-
bonding acyl ligand and one weakly σ-bonding methanol
ligand, and the relative thermodynamic stabilities of the
isomers will depend on a balance between the metal-ligand
bonding and steric congestion. In order to probe this further
and thereby gain insight into the factors that make L3 such an
effective ligand for the hydromethoxycarbonylation of
(13) (a) Eastham, G. R.; Heaton, B. T.; Iggo, J. A.; Tooze, R. P.;
Whyman, R.; Zacchini, S. Chem. Commun. 2000, 609. (b) Clegg, W.;
Eastham, G. R.; Elsegood, M. R. J.; Heaton, B. T.; Iggo, J. A.; Tooze, R. P.;
Whyman, R.; Zacchini, S. Organometallics 2002, 21, 1832. (c) Clegg, W.;
Eastham, G. R.; Elsegood, M. R. J.; Heaton, B. T.; Iggo, J. A.; Tooze, R. P.;
Whyman, R.; Zacchini, S. J. Chem. Soc., Dalton Trans. 2002, 3300. (d)
Eastham, G. R.; Tooze, R. P.; Kilner, M.; Foster, D. F.; Cole-Hamilton, D. J.
J. Chem. Soc., Dalton Trans. 2002, 1613.
(14) (a) Pugh, R. I.; Drent, E. Adv. Synth. Catal. 2002, 344, 837. (b)
van Leeuwen, P. W. N. M. In Modern Carbonylation Methods; Kollar, L.,
Ed.; Wiley: New York, 2008. (c) Suykerbuyk, J. C. L. J.; Drent, E.; Pringle,
P. G. (Shell) World Patent WO 98/42717, 1998. (d) Zuidema, E.; Bo, C.;
van Leeuwen, P. W. N. M. J. Am. Chem. Soc. 2007, 129, 3989.
(15) (a) Liu, J.; Heaton, B. T.; Iggo, J. A.; Whyman, R. Chem.
Commun. 2004, 1326. (b) Donald, S. M. A.; Macgregor, S. A.; Settels, V.;
Cole-Hamilton, D. J.; Eastham, G. R. Chem. Commun. 2007, 562.
(16) (a) Dekker, G. P. C. D.; Elsevier, C. J.; Vrieze, K.; van Leeuwen,
P. W. N. M. Organometallics 1992, 11, 1598. (b) Ledford, J.; Shultz, C. S.;
Gates, D. P.; White, P. S.; DeSimone, J. M.; Brookhart, M. Organometallics
2001, 20, 5266.
(17) Koide, Y.; Bott, S. G.; Barron, A. R. Organometallics 1996, 15,
2213.
(18) Freixa, Z.; van Leeuwen, P. W. N. M. Dalton Trans. 2003, 1890.

Article
Organometallics, Vol. 29, No. 10, 2010 2295
Scheme 4
Table 3. 31P NMR Dataa
PBu2 signal
PPh2 signal
compound
δP
1JPtP
δP
1JPtP
2JPP
Figure 1. Thermal ellipsoid plot of the structure of [PtCl2(L3)]
(3). Hydrogen atoms and the solvate CH2Cl2 have been omitted
for clarity. Selected bond lengths (A) and angles (deg): Pt(1)-
P(1) 2.2935(8), Pt(1)-P(2) 2.2489(8), Pt(1)-Cl(1) 2.3628(8) Pt-
(1)-Cl(2) 2.3669(8), P(1)-C(1) 1.862(3), P(1)-C(21) 1.908(3),
P(1)-C(25) 1.883(3), P(2)-C(8) 1.842(3), P(2)-C(9)1.828(3),
P(2)-C(15)1.824(3), P(1)-Pt(1)-P(2) 100.51(3), Cl(1)-Pt(1)-
Cl(2) 83.99(3), P(1)-Pt(1)-Cl(1) 93.23(3), P(2)-Pt(1)-Cl(2)
82.29(3).
1
2
3
4
5
6
7
11.9
3644
-1.8
-7.0
3571
3606
25.0
19.3
3603
1853
13
6.4
5.4
1913
1929
27.6
19.5
24.0
1883
1773
4384
15
13
8
7.3
8.9
7.4
4.3
1773
4350
4532
1704
15
9a
9b
31.3
31.4
1828
4267
15
16
a Data recorded in CH2Cl2 or CD2Cl2.
ethene, the coordination chemistry of L3 with Pd and Pt was
investigated. The chemistry of hydride, methyl, and acyl
complexes of L3 were particularly targeted as potential
models of the catalytic intermediates shown in Scheme 3.
Platinum Coordination Chemistry. The platinum chelates
1-3 and 4-6 were made according to the route shown in
Scheme 4, and the 31P NMR parameters are given in Table 3.
Crystals of 3 as its CH2Cl2 solvate were grown from
CH2Cl2/hexane, and the crystal structure geometry is shown
in Figure 1. The structure determination reveals that the
PtBu2 donor has an essentially identical trans influence to
PPh2; the Pt-Cl bond trans to the PtBu2 group is only
ca. 0.004 A longer than that trans to the PPh2 group.
The Pt-PtBu2 bond is significantly longer (ca. 0.045 A) than
t
the Pt-PPh2, reflecting the greater steric bulk of the Bu
groups.
1
The 31P and H NMR spectra were recorded for 3 from
-90 to þ100 °C. No change was apparent in the 31P NMR
spectra, but in the 1H NMR spectra, fluxionality was evident
Figure 2. Variable-temperature 1H NMR for the tBu region of 3
(a) in C2D2Cl4 (above 20 °C) and (b) in CD2Cl2 (below 20 °C).
t
from the signals for the Ph and Bu groups (Figure 2). At
t
room temperature, two broad Bu signals are observed
consistent with the chelate ring adopting the envelope con-
formation detected in the X-ray crystal structure of 3
(Figure 1), which makes the tBu groups inequivalent
(labeled R and β in Scheme 5; see C21 and C25 in
Figure 1). At 100 °C, ring inversion is sufficiently rapid on
t
Between 0 and -60 °C, the two Bu doublets sharpen.
t
Below -60 °C, one of the Bu signals collapses, and this is
interpreted as being due to restricted C-P rotation in the
R-tBu (due to interaction with the phenylene backbone),
making each of the methyls in the R-tBu group inequiva-
lent.19 Broad 1H NMR signals at ambient temperatures were
t
the NMR time scale that the Bu groups (R and β) appear
as a single doublet (see Figure 2).

2296 Organometallics, Vol. 29, No. 10, 2010
Fanjul et al.
Scheme 5
led to brown solutions containing complex mixtures of
products in solution according to 31P NMR analysis (see
below for the calculated structures of 11a/11b). The cationic
hydridoplatinum(II) complex 12b was characterized in solu-
tion as a single isomer upon treatment of [PtH(Cl)(PPh3)2]
with L3, but attempts to isolate 12b led to a dark brown oil
that contained several unidentified products of decomposi-
tion. The geometry of 12b in solution is based on its
characteristic 31P NMR spectrum, which showed that the
PPh2 signal was coupled strongly to the hydride (JPH of 158
Hz), consistent with a trans Ph2P-Pt-H arrangement.
also observed for many of the complexes reported below (see
Experimental Section), and it was assumed that conforma-
tional fluxionality of the type shown in Scheme 5 is respon-
sible, though no further variable-temperature NMR studies
have been carried out.
The chelate complexes [PtCl(CH3)(L1)] (7) and [PtCl(CH3)-
(L2)] (8) were generated in solution by treatment of [PtCl(CH3)-
(cod)] with L1 or L2, respectively, and characterized by 31P
NMR only (see Table 3). Treatment of [PtCl(CH3)(cod)] with
L3 gave a mixture of two isomers (9a and 9b) inan initial ratioof
ca. 200:1, which gradually changed over 4 days to a ratio of
ca. 90:1. The same species were formed upon treatment of 6with
ethereal HCl (Scheme 6), initially in a ca. 60:1 ratio of isomers 9a
and 9b which increased over 8 days to ca. 90:1. It was concluded
that 9a and 9b slowly interconvert and that 90:1 is the equilib-
rium ratio (eq 1).
Palladium(II) Coordination Chemistry. Treatment of
[PdCl2(NCPh)2] with L3 gave the chelate 13, which has been
fully characterized (see Experimental Section). Treatment of
[PdCl(CH3)(cod)] with L3 gave a 20:1 mixture of isomers
assigned the structures 14a and 14b, by analogy with the
platinum chemistry.
The structure of the major product 9a was unambiguously
assigned as having the CH3 group cis to the PPh2 donor on
the basis of the 31P NMR parameters (Table 3) and 1H NOE
spectroscopy, which showed an intensity increase of some of
the aromatic signals upon irradiation of the Pt-CH3 signal.
It is notable that the values of JPtP are significantly higher in
9a (and lower in 9b) than the corresponding signals in the
symmetrical analogues 7 and 8 (see Table 3), which perhaps
indicates a mutual strengthening of the Pt-P bonds in the
more stable isomer 9a (see the results from the computa-
tional study below).
Treatment of the 9a/9b mixture with 1 equiv of PMe3 gave
quantitative conversion to the cationic species 10b, and the
isomer 10a was not detected, leading to an estimate of the
ratio of 10b:10a of at least 50:1.20 Thus the stereochemistry of
the major isomer in this reaction (eq 2) has been inverted.
Crystals of 14a were grown from CH2Cl2/hexane and
shown to be representative of the major isomer by redissol-
ving them in CDCl3 and measuring the 31P NMR spectrum,
which showed that isomer 14a was present to the extent of at
least 95%. The crystal structure (see Figure 3) confirms that
14a has the CH3 group trans to the tBu2P donor and therefore
has the same geometry as the major isomer of the platinum
analogue 9a. The Pd-PtBu2 bond is ca. 0.178 A longer than the
Pd-PPh2 bond, in line with the greater steric bulk of the PtBu2
donor and the high trans influence of the methyl group. The
disorder in 14a shows Cl(2) with approximately 3.5% occu-
pancy. Unfortunately the corresponding carbon peak was not
observed in the difference map (due to the low occupancy), and
therefore it is inconclusive (from the X-ray analysis) whether the
low occupied fraction is 14b or the dichloro complex 13.
Attempts to make the neutral [PtCl(H)(L3)] (11a/11b) by
reduction of 3 with NaBH4, LiBHEt3, or hydrazine hydrate
(19) The aromatic 1H NMR signals (6.8-8.0 ppm) are also broad at
ambient temperatures and change with temperature. However the
complexity due to overlapping resonances in this region precluded any
detailed analysis.
(20) Complex 10b was characterized in solution only by 31P NMR
spectroscopy: tBu2P, δ 25.6 (dd, JPP 368 and 24 Hz, JPPt 2583 Hz); Ph2P,
-2.5 (t, JPP 24 Hz, JPPt 1963 Hz); PMe3, -18.8 (dd, JPP 368 Hz, JPP 24
Hz, JPPt 2520 Hz).

Article
Organometallics, Vol. 29, No. 10, 2010 2297
Figure 4. Thermal ellipsoid plot of the structure of [PdH(PCy3)-
(L3)][BF4] (15b). The [BF4] cation and the hydrogen atoms, except
H(100), have been omitted for clarity. Selected bond lengths (A)
and angles (deg): Pd(1)-P(1) 2.3293(14), Pd(1)-P(2) 2.3619(14),
Pd(1)-P(3) 2.3322(13), Pd(1)-H(100) 1.53(8), P(1)-C(1)
1.853(5), P(1)-C(21) 1.886(6), P(1)-C(25) 1.892(5), P(2)-C(8)
1.853(5), P(2)-C(9) 1.813(5), P(2)-C(15) 1.828(5), P(1)-
Pd(1)-P(2) 103.13(5), P(2)-Pd(1)-P(3) 104.72(5), P(3)-
Pd(1)-H(100) 73(3), P(1)-Pd(1)-H(100) 79(3).
Figure 3. Thermal ellipsoid plot of the structure of [PdCl(CH3)-
(L3)] (14a). Hydrogen atoms have been omitted for clarity.
Selected bond lengths (A) and angles (deg): Pd(1)-P(1)
2.4051(8), Pd(1)-P(2) 2.2274(7), Pd(1)-Cl(1) 2.3824(7), Pd(1)-
Cl(2) 2.3824(7), Pd(1)-C(29) 2.119(2), P(1)-C(1) 1.872(2),
P(1)-C(21) 1.906(2), P(1)-C(25) 1.876(2), P(2)-C(8) 1.858(2),
P(2)-C(9) 1.820(2), P(2)-C(15) 1.831(2), P(1)-Pd(1)-
P(2) 101.21(3), Cl(1)-Pd(1)-C(29) 83.35(6), Cl(1)-Pd(1)-
Cl(2) 84.9(12), P(1)-Pd(1)-Cl(1) 92.87(3), P(2)-Pd(1)-C(29)
82.59(6).
The cationic hydridopalladium complex 15b was made by
treatment of [Pd(PCy3)2]21 with L3 followed by addition of
ethereal HBF4. The solution structure of 15b with the H cis to
the tBu2P group was assigned as a single isomer on the basis
of 31P and 1H NMR spectroscopy (see Experimental
Section). The crystal structure of 15b (see Figure 4) confirms
the isomer assignment and shows angular distortion from
ideal square-planar geometry as a consequence of the com-
bination of steric bulk of the phosphine ligands and the large
bite angle of L3. A similar distortion has been observed in
[PdH(iPr2P(CH2)3PiPr2)(PCy3)][OTf].22
Scheme 6
Stereoelectronic Effects in Determining Geometric Isomer
Preference. The complexes [MCl(CH3)(L3)] where M = Pt
(9) or Pd (14) resemble critical intermediates in the carbony-
lation cycle (Scheme 3) in that they contain one strong
σ-donor and one relatively weak σ-donor ligand. In both
cases, the a isomer (with CH3 trans to PtBu2) predominates
over the b isomer. From the equilibrium ratios, the energy
differences between the isomers are small (<3 kcal mol-1
)
but such subtle differences in the stability of isomeric Pd
catalytic intermediates may be important to the success of
the carbonylation catalysts (see below). Therefore, it is worth
considering the factors involved in determining the relative
isomer stabilities.
According to the antisymbiotic effect,23-25 ligands with a
high trans influence prefer to be trans to ligands of low trans
(21) Yoshida, T.; Otsuka, S. Inorg. Synth. 1990, 28, 113.
(22) CSD refcode CULLIE. Perez, P. J.; Calabrese, J. C.; Bunel, E. E.
Organometallics 2001, 20, 337.
(23) (a) Pearson, R. G. Inorg. Chem. 1973, 12, 712. (b) Crabtree, R. H.;
Davis, M. W.; Mellea, M. F.; Mihelcic, J. M. Inorg. Chim. Acta 1983, 72,
223. (c) Navarro, R.; Urriolabeitia, E. P. J. Chem. Soc., Dalton Trans. 1999,
4111. (d) Battan, L.; Fantasia, S.; Manassero, M.; Pasini, A.; Sansoni, M.
Inorg. Chim. Acta 2005, 358, 555. (e) Canac, Y.; Debono, N.; Vendier, L.;
Chauvin, R. Inorg. Chem. 2009, 48, 5562. (f) Djukic, J.-P. In Palladacycles.
Synthesis, Characterization and Applications; Dupont, J.; Pfeffer, M., Eds.;
Wiley-VCH: Weinheim, Germany, 2008.
Constraints were required to model the disorder. The
Pd(1)-Cl(2) bond length was constrained to match the
Pd(1)-Cl(1) bond length, as were the atomic displacement
parameters.

2298 Organometallics, Vol. 29, No. 10, 2010
Fanjul et al.
influence. This has been used to explain why, in complexes of
the type [MCl(CH3)(PR3)2], the trans isomer (with CH3 trans
to Cl) is strongly favored26 and in the ditopic chelates
[PdCl(Ar){Ph2PCH2CH2(2-pyridyl)}], the Ar group is trans
to the pyridine donor.27 The rationale for the antisymbiotic
effect is that trans ligands are in competition for the same sd
hybrid orbital on the metal involved in σ-bonding to the
ligands.24,25 On the basis of the antisymbiotic effect, isomers
9a and 14a, with the strong σ-donating CH3 trans to the more
strongly σ-donating tBu2P-donor, would be predicted to be
less stable than 9b and 14b. The fact that the opposite is
observed shows that antisymbiosis alone does not determine
the isomer preference.28,29 Others have also found28,30 that in
chelates of ditopic diphosphorus ligands of the type [MCl-
(CH3)(diphos)] (M = Pd or Pt), the preferred isomer is not the
one predicted from considerations of antisymbiosis.
Scheme 7
It seems likely that the steric bulk of the group is a factor
contributing to the observed a/b isomer stabilities. Higher
steric demands of ligand X than CH3 would destabilize
isomer a relative to b due to clashing with the adjacent PtBu2
group (see Chart 1, R = CH3). The greater van der Waals
volume of CH3 (ca. 13.7 cm3 mol-1) than Cl (ca. 11.6 cm3
longer than the Pt-Cl distance trans to the tBu2P donor; not
unexpectedly, the Pt-Cl distances trans to the Ph2P donor in
16 and 3 are very similar. (b) The greater steric bulk of the
tBu2P compared to nBu2P is evident from the difference in the
Bu2P(1)-Pt-Cl(1) and Ph2P(2)-Pt-Cl(2) angles; in 3 these
angles differ significantly at ca. 93° and 82°, whereas in 16 the
corresponding angles are very similar to each other (ca. 86°
and 87°). The crystallography is consistent with ligand L4
being characterized as a less bulky analogue of L3; that is, the
nBu2P is at least as good a donor as tBu2P but is considerably
less bulky.
mol-1 31
) would contribute to the greater stability of 9a than
9b. Steric effects can also explain the preference for the b
geometry of [Pt(CH3)(PMe3)(L3)]Cl (10b) and the hydride
complexes [PtH(PPh3)(L3)]Cl (12b) and [PdH(PCy3)(L3)]Cl
(15b) since in each case the bulky PR3 occupies the site trans
to the tBu2P group.32
Treatment of [PtCl(CH3)(cod)] with L4 gave, after 24 h,
the products 17a and 17b (Scheme 7) in the ratio of ca. 2:1.33
Therefore the effect of substituting the PtBu2 in L3 for PnBu2
in L4 has led to a reduction in the a:b isomer ratio from 90:1
(for 9a:9b) to 2:1 (for 17a:17b). This result is consistent with
there being a major steric component to the explanation of
the a:b isomer ratio.
The conclusion from the synthetic studies is that the
isomer ratios for [MCl(CH3)(L3)], [MH(PR3)(L3)]Cl (M =
Pd or Pt), [Pt(CH3)(PR3)(L3)]Cl, and [PtCl(CH3)(L4)] are
controlled mainly by steric factors. The antisymbiotic effect
appears to play a minor role in determining the stability of
complexes of L3, perhaps because the difference between the
P-donors in terms of their σ-bonding properties is small. It
follows that the dominant effect is steric and associated with
the bulky PtBu2 creating congestion at the sites adjacent to it.
It was hoped that a deeper understanding of this system
would emerge from computational analysis of the isomer
preferences.
Computational Analysis. It is difficult to derive a full
understanding of the observed isomer preferences from the
experimental data alone, especially since not all complexes
have been amenable to crystallographic analysis, and hence
the structural data are incomplete. We have thus used density
functional theory (DFT) calculations to generate a more
complete structural and energetic data set, exploring further
substituents, and to test our theories (see Experimental
A factor that may be important in determining the isomer
preference in [MCl(CH3)(L3)] is the stereoelectronic conse-
quence of the interaction of the bulky tBu2P group with the
adjacent PPh2 group. To probe this further, the coordination
chemistry of ligand L4 (Scheme 7) was investigated. It was
reasoned that the nBu2P group in L4 would have similar
σ-donor properties to the tBu2P in L3 but much lower steric
bulk.11 Ligand L4 was prepared by a similar route to that
used for L3 (Scheme 1) and fully characterized (see Experi-
mental Section). The similarity of the JPSe values for the
diselenide L4(Se)2 (687 and 729 Hz) to those for L3(Se)2 (694
and 724 Hz) is consistent with L4 and L3 having similar
σ-donor ability. Treatment of [PtCl2(cod)] with L4 gave the
chelate 16, for which crystals suitable for X-ray crystal-
lography were grown.
From the crystal structure data (see Figure 5) for 16 and 3,
the following can be deduced. (a) The trans influence of the
nBu2P donor in 16 is slightly greater than the tBu2P donor in
3 since the Pt-Cl distance trans to the nBu2P donor is ca. 0.02 A
(24) Davies, J. A.; Hartley, F. R. Chem. Rev. 1981, 81, 79.
(25) Harvey, J. N.; Heslop, K. M.; Orpen, A. G.; Pringle, P. G. Chem.
Commun. 2003, 278.
(26) Allen, F. H.; Pidcock, A. J. Chem. Soc. A 1968, 2700.
(27) Casares, J. A.; Espinet, P.; Soulantica, K.; Pascual, I.; Orpen, A.
G. Inorg. Chem. 1997, 36, 5251.
(28) Leone, A.; Gischig, S.; Consiglio, G. J. Organomet. Chem. 2006,
691, 4816.
(29) (a) Brunker, T. J.; Blank, N. F.; Moncarz, J. R.; Scriban, C.;
Anderson, B. J.; Glueck, D. S.; Zakharov, L. N.; Golen, J. A.; Sommer,
R. D.; Incarvito, C. D.; Rheingold, A. L. Organometallics 2005, 24, 2730.
(b) Roy, A. H.; Hartwig, J. F. J. Am. Chem. Soc. 2003, 125, 8704. (c) Roy, A.
H.; Hartwig, J. F. Organometallics 2004, 23, 194.
(30) (a) Nozaki, K.; Sato, N.; Tonomura, Y.; Yasutomi, M.; Takaya,
H.; Hiyama, T.; Matsubara, T.; Koga, N. J. Am. Chem. Soc. 1997, 119,
12779. (b) Gambs, C.; Consiglio, G.; Togni, A. Helv. Chim. Acta 2001, 84,
3105.
(33) After 1 h, a 1:12 mixture of the isomers 14a and 14b was
observed, which shows that 14b is favored kinetically. This is consistent
with the more nucleophilic nBu2P donor initally substituting the Cl to
give [Pt(CH3)(η1-nBu2PCH2C6H4PPh2)(cod)]Cl followed by chelation
(putting PPh2 trans to CH3) and then substitution of cod. See: (a) Goel,
A. B.; Goel, S. J. Indian Chem. Soc. 1982, 59, 141. (b) Mikhel, I. S.;
Bondarev, O. G.; Tsarev, V. N.; Grintselev-Knyazev, G. V.; Lyssenko,
K. A.; Davankov, V. A.; Gavrilov, K. N. Organometallics 2003, 22, 925.
(c) Seibert, M.; Merzweiler, K.; Wagner, C.; Weichmann, H. J. Orga-
nomet. Chem. 2003, 687, 131.
(31) Bondi, A. J. Phys. Chem. 1964, 68, 441.
(32) The preference for the b stereochemistry for the hydrides 12 and
15 would also be favoured by antisymbiosis.

Article
Organometallics, Vol. 29, No. 10, 2010 2299
Chart 3
from the 90:1 ratio). Less good agreement was found for
complexes 17a/17b, where the calculated energy difference
was -1.0 kcal mol-1, while the experimental value is þ0.41
kcal mol-1 (calculated from the 2:1 ratio); however, the calcu-
lated energy difference between the isomers 17a/17b is very
sensitive to the chosen methodology, ranging from -1.0 to þ
0.3 kcal mol-1 for different functionals (see Supporting Infor-
mation for details).
A larger discrepancy was found for the palladium com-
plexes 14a/14b, where the calculated energy difference of
4.2 kcal mol-1 is significantly greater than the experimental
value of 1.8 kcal mol-1. This calculated energy difference for
14a/14b ranges from 3.8 to 4.2 kcal mol-1 for different
functionals, but even the lowest value of 3.8 kcal mol-1 is
still over 2 kcal mol-1 from the experimental value. Thus the
experimentally observed isomer energy difference is not
reproduced as well for the palladium species, but the correct
preference for isomer a has again been captured.
Figure 5. Thermal ellipsoid plot of the structure of [PtCl2(L4)]
(16). Hydrogen atoms, a second complex, and two molecules of
CH2Cl2 have been omitted for clarity. Selected bond lengths (A)
and angles (deg): Pt(1)-P(1) 2.250(2), Pt(1)-P(2) 2.240(2), Pt-
(1)-Cl(1) 2.348(2), Pt(1)-Cl(2) 2.385(2) P(1)-C(1) 1.829(11),
P(1)-C(21) 1.806(12), P(1)-C(25) 1.793(11), P(2)-C(8)
1.813(9), P(2)-C(9) 1.838(11), P(2)-C(15) 1.819(10), P(1)-
Pt(1)-P(2) 99.60(8), Cl(1)-Pt(1)-Cl(2) 87.87(8), P(1)-
Pt(1)-Cl(1) 86.06(8), P(2)-Pt(1)-Cl(2) 86.37(8).
Section for details of the computational method used and the
Supporting Information for a discussion of method effects).
The structures of the complexes shown in Chart 3 have
been optimized, and all of the computational results are
summarized in Table 4. Complexes 11, 18, and 19 have not
been synthesized experimentally.
The calculations on complexes 9 and 17-19 allow an
assessment of the importance of steric effects. The energetic
preference for isomer a over isomer b increases with in-
creasing P-substituent size11 (PMe2 < PnBu2 < PCy2 <
PtBu2). The increase on going from PCy2 to PtBu2 is consider-
able, suggesting that the relationship is not linear. However, the
change in donor strength in this series of ligands can also affect
the isomer preferences, and indeed the calculated differences
(albeit small) in the Pt-Y bond lengths (trans to the PR2
groups) suggest that this merits consideration.
Complexes 11a/11b were not accessible experimentally but
were calculated in order to explore the effect of replacing the
CH3 with an H ligand, which is smaller than CH3 but has a
similarly high trans influence. The calculated structures of
11a and 11b reveal shorter Pt-P distances trans to the
hydride compared to Pt-P bond lengths trans to methyl in
complexes 9a and 9b, indicating that H has a lower trans
influence than CH3. A switch in isomer preference was
observed with 11b being 3.1 kcal mol-1 lower in energy than
11a, once again consistent with steric effects being decisive in
determining the isomer ratio in these systems.
As expected for DFT calculations with a gradient-
corrected functional such as PW91 used here, bond lengths
are slightly longer than those observed crystallographi-
cally,34 but show similar trends. The calculated Pt-P bond
lengths for the favored isomer 9a (2.444 and 2.249 A) are
both shorter than the analogous Pt-P bond lengths in the
symmetrical species 7 (2.473 A) and 8 (2.252 A).35 These data
support the conclusion from the JPtP values (see above) that
both of the P-donors in 9a are more strongly bound to the Pt
than the analogous P-donor in the complexes 7 and 8; that is,
the donors mutually strengthen the Pt-P bonding in 9a. This
synergic effect on the bonding may be part of the explanation
for the greater stability of the Pd-L3 catalyst than the
Pd-L1 catalyst described above (Table 3).
The calculations confirm the observed isomer preferences
and are in reasonably good agreement with experimentally
observed energy differences calculated from the isomer ratios
for the platinum complexes. For example, the energy difference
of 3.1 kcal mol-1 that was calculated for complexes 9a/9b is
close to the experimental value of 2.7 kcal mol-1 (calculated
A noteworthy observation from the calculated structures
shown in Table 4 was the presence of several short contacts
between the P-donor substituents and the methyl and chlor-
ide ligands in both isomers. For the platinum complex 9a, we
(34) (a) Fey, N.; Harris, S. E.; Harvey, J. N.; Orpen, A. G. J. Chem.
Inf. Model. 2006, 46, 912, and references therein. (b) Kapoor, P. N.; Kakkar,
R. J. Mol. Struct. (THEOCHEM) 2004, 679, 149.
(35) The converse is also true: the calculated Pt-P bond lengths for
the disfavored isomer 9b (2.300 and 2.441 A) are both longer than the
analogous Pt-P bond lengths in the symmetrical species 7 (2.287 A) and
8 (2.252 A).
observed a short Cl H contact (2.497 A calcd, 2.563 A in
3 3 3
XRD) to one of the tBu groups, but in 9b, several very short

2300 Organometallics, Vol. 29, No. 10, 2010
Table 4. Calculated Isomer Preferences and Key Structural Parameters of Complexes Studieda
Fanjul et al.
complex
ΔE (kcal mol-1
)
M-PPh2 (A)
M-PR2 (A)
M-Xb (A)
M-Yb (A)
P-M-P (deg)
3
7
8
9a
9b
2.295 (2.249)
2.287c
2.252
2.249
2.411
2.338 (2.294)
2.473
2.394
2.444
2.300
2.416 (2.363)
2.431
2.409
2.437
2.106
2.437 (2.367)
2.110
2.112
2.115
2.449
100.9 (100.5)
103.0
101.3
101.2
100.2
0.00
3.05
11a
11b
14a
14b
16
17a
17b
18a
18b
19a
19b
0.00
2.248
2.377
2.270 (2.227)
2.441
2.291 (2.231)
2.252
2.393
2.269
2.281
2.427
2.271
2.476 (2.405)
2.333
2.295 (2.242)
2.393
2.265
2.413
2.415
2.430
1.610
2.441 (2.382)
2.095
2.418 (2.361)
2.434
2.106
2.428
2.436
1.610
2.437
2.105 (2.119)
2.444
2.430 (2.383)
2.111
2.434
2.102
2.094
103.3
103.4
101.4 (101.2)
100.6
101.0 (99.3)
102.3
98.0
102.1
100.6
-3.06
0.00
4.15
0.00
-1.03
0.00
-0.92
0.00
0.97
2.247
2.399
2.376
2.255
2.439
2.106
2.110
2.438
102.1
101.2
a Crystallographic data are given in parentheses if available. b M-X is trans to the PPh2 donor, and M-Y is trans to the PR2 donor. c Pt-PtBu2
distance.
contacts can be identified between the CH3 ligand and both
and 13C NMR spectra and IR absorptions at 2052 and
1677 cm-1. Over the following 6 days the proportion of
20a and 21a increased to 8% and 40%, respectively. The
structure of 20a was supported by the following independent
synthesis. Treatment of a CH2Cl2/MeCN (1:1) solution of
9a/9b with AgBF4 gave the cationic species 22a with the
MeCN cis to the PtBu2 group (the geometric isomer of 22a
was not detected), which then reacted rapidly with 13CO to
give 20a0, which had almost identical 31P and 13C NMR
parameters to the chloride salt 20a (Scheme 8).
of the tBu groups (shortest: 2.000, 2.253 A). A similar pattern
was observed in the corresponding palladium complexes,36
where even shorter contacts in 14b may be contributing to the
(anomalously) large calculated energetic preference. Short
contacts were also detected between the C-H groups in
P-Ph and P-nBu substituents and Pt-Cl and Pt-CH3
ligands in complexes 9, 14, and 17, but these were all
significantly longer than the contacts with the P-tBu sub-
stituents discussed above.
With the proviso that the calculated energy differences are
close to the limits of computational accuracy, the main
conclusion from the calculations is that the origin of the
energy differences between the a and b isomers (Chart 3)
is predominantly steric. This can be amplified in the
particular case of the isomers of [MCl(CH3)(L3)] since very
The sequence of reactions shown in Scheme 9 is proposed
to explain the formation of the observed products 20a and
21a. Complex 20a is the result of displacement of Cl from
major isomer 9a. Complex 20b is the predicted product of
displacement of Cl from minor isomer 9b, which upon
methyl migration to CO would give 21a.
short, repulsive H H contacts between the adjacent
3 3 3
The formation of acyl complex 21a could result from any
of the three pathways shown in Scheme 9 (M = Pt): (a) slow
isomerization of 9a to 9b followed by displacement of Cl by
CO to give 20b and then rapid migration of CH3 to CO
CH3 and PtBu2 groups are detected in the less favored b
isomers.
Extension of these calculations to the evaluation of com-
plexes considered to lie in the catalytic cycle (Scheme 3) is
complicated by the increased conformational complexity
introduced by coordination of acyl and methanol ligands,
which requires extensive conformational searches and
potentially increases the noise levels in calculated isomer
preferences. Nevertheless, work is currently underway to
explore the effects of using ligands such as L1-4 on the
isomer preferences and reactivity in key steps of the catalytic
cycle.
Model Carbonylation Studies. In order to probe the source
of the MeP selectivity of the Pd-L3 catalyst, model methyl-
platinum complexes 9a/9b and methyl-palladium complex
14a/14b were treated with labeled 13CO, and the products
were unambiguously characterized by 31P and 13C NMR
spectroscopy (see Experimental Section for these data).
A CH2Cl2 solution of 9a/9b was stirred under an atmo-
sphere of 13CO at room temperature. After 3 h, apart from
starting materials 9a/9b, two minor products were detected,
which were assigned to terminal carbonyl species 20a (2%)
and the acyl complex 21a (10%) from the characteristic 31P
t
promoted by the bulky, adjacent Bu2P group; (b) slow
isomerization of 20a to 20b and then rapid migration of
CH3 to CO; (c) slow migration of CH3 to CO in 20a to give
21b followed by rapid equilibration to 21a. Pathway (a) is
viable, as the slow isomerization of 9a to 9b has been
observed (see discussion of eq 1 above). The BF4 salt 20a0
(Scheme 8) showed no evidence for products of CH3 migra-
tion after several days in solution, and refluxing a CH2Cl2/
MeCN solution of 20a0 for 4 h led to substantial conversion
back to MeCN complex 22a. Addition of 1 equiv of [NBu4]Cl
to 20a0 led to quantitative formation of 9a/9b, and thus no
supporting evidence was obtained for pathways (b) and (c).
When a CH2Cl2 solution of a 20:1 mixture of methyl-
palladium complexes 14a/14b was stirred under an atmo-
sphere of 13CO, a rapid reaction took place to give a ca. 10:1
mixture of the acyl complexes 23a/23b quantitatively (νCO 1655
cm-1); the terminal CO complexes 24a/24b (Scheme 9),
which are the presumed intermediates, were not observed.
From the 31P and 13C NMR spectra, it was deduced that the
predominant complex 23a has the acyl trans to the PtBu2
group (δ(PtBu2) 27.7, JPC 103 Hz, JPP 61 Hz; δ(PPh2) 15.5,
JPC 14 Hz, JPP 61 Hz). Treatment of 23a/23b with MeOH gave
Me13CO2Me immediately, as shown by the prominent
13C NMR signal at δC 176.6, and a complex hydride signal
(36) Short contacts detected in the following Pd complexes. 14a:
HtBu Cl 2.511 A, HMe HPh 2.598 A; 14b: HPh Cl 2.644 A,
3 3 3
3 3 3
3 3 3
3 3 3
HMe
H
tBu 1.968, 2.032, 2.228 A; 17a: HnBu Cl 2.720 A, HMe HPh
HnBu 2.171, 2.260, 2.314 A.
3 3 3 3 3 3
2.522 A; 17b: HPh Cl 2.727 A, HMe
3 3 3
3 3 3

Article
Organometallics, Vol. 29, No. 10, 2010 2301
Scheme 8
Scheme 9
was observed in the 1H NMR spectrum at -6.94 ppm. Com-
plex 23a is an analogue of the MeP-producing intermediate A1
(Chart 2).
Scheme 10
The potential pathways to 23a/23b upon treatment of 14a/
14b with CO are denoted (a)-(c) in Scheme 9 (M = Pd) and
are analogous to those given above for the related, much
slower, platinum chemistry in Scheme 9 (M = Pt). The
feasibility of pathway (c) involving the rapid isomerization of
the acyl products 23a/23b is supported by the observations of
Iggo et al.37 on the symmetrical diphosphine complex [Pd-
(COEt)(solv)(L1)]BF4, which undergoes acyl migration be-
tween the degenerate cis sites very rapidly on the NMR time
scale at ambient temperatures.
The model studies (Schemes 9, M = Pt or Pd) and the
calculated isomer stabilities are consistent with the explana-
tion of the MeP selectivity of the Pd-L3 catalyst being a
combination of thermodynamic and kinetic factors depicted
in Scheme 10. It is suggested that the product-determining
(37) Clegg, W.; Eastham, G. R.; Elsegood, M. R. J.; Heaton, B. T.;
Iggo, J. A.; Tooze, R. P.; Whyman, R.; Zacchini, S. Organometallics
2002, 21, 1832.

2302 Organometallics, Vol. 29, No. 10, 2010
Fanjul et al.
acyl intermediates A1 and A2 are in dynamic equilibrium with
the MeP-producing acyl intermediate A1 predominating due
to its greater thermodynamic stability. The sites occupied by
MeOH in A1 and A2 resemble the sites occupied by MeOH in
the analogous acyl intermediates (shown in Scheme 10)
proposed for the Pd-L1 catalyst (which gives MeP rapidly)
and the Pd-L2 catalyst (which gives oligomers and PK
slowly). The combination of lower stability and lower re-
activity of A2 than A1 would explain the MeP selectivity of
Pd-L3 catalysts.
BH3 from Lucite International. All phosphines were stored
under nitrogen at room temperature. All other reagents were
used as received from Aldrich, Strem, or Lancaster. NMR
spectra were recorded on a Jeol Delta 270, Jeol Eclipse 300,
Jeol Eclipse 400, Varian 400, or Lambda 300. Infrared spectro-
scopy was carried out on a Perkin-Elmer 1600 Series FTIR.
Mass spectra were recorded on a MD800 by the Mass Spectro-
metry Service, University of Bristol. Elemental analyses were
carried out by the Microanalytical Laboratory of the School of
Chemistry, University of Bristol.
Synthesis of r-(Di-tert-butylphosphino)-r0-(diphenylphosphino)-
xylene (L3). A 1.6 M solution of BuLi (7.60 mL, 11.8 mmol)
in hexane was added to a solution of Bu2PH BH3 (1.60 g,
t
Conclusions
3
10.0 mmol) in THF (15 mL) at -78 °C. The mixture was allowed
to warm slowly to room temperature, stirred for 30 min at this
The hydromethoxycarbonylation of ethene to give methyl
propanoate catalyzed by Pd complexes of bulky ditertiary
phosphines (the Lucite Process) is one of the most elegant
applications of homogeneous catalysis in the commodity
chemicals business. The commercial success of the Lucite
Process is predicated on the rate, selectivity, and stability of
the Pd-L1 catalyst. It was reasonable to assume that the
presence of two bulky PR2 (almost invariably PtBu2) groups
was crucial to the efficiency of the catalyst. However, we
have shown here, from the results with the Pd-L3 catalyst,
that only one PtBu2 is required to obtain excellent selectivity
and high activity. Advantageously, the new Pd-L3 catalyst
is also considerably more stable than the commercial Pd-L1
catalyst under the same conditions. The unsymmetrical
ligand L3 is representative of a new class of ligands for alkene
hydromethoxycarbonylation catalysis and opens the way to
the development of other ditopic ligands for this catalysis.
An insight into why L3 is such an effective ligand for the
hydromethoxycarbonylation catalysis comes from its Pt and
Pd coordination chemistry. The bulk of the PtBu2 donor
controls the stereochemistry of the complexes of L3 with the
important consequence that the critical intermediate is pre-
dicted to have the propanoyl ligand trans to the PtBu2 donor
(A1 in Scheme 10), making it resemble the exclusively MeP-
producing intermediate with the Pd-L1 system. The source
of the higher stability of the Pd-L3 catalyst than the Pd-L1
catalyst stability is worthy of further investigation but may
be associated with the mutual strengthening of the Pd-P
bonds in the Pd-L3 complexes.
temperature, and recooled to -78 °C, and then the tBu2PLi BH3
3
solution was added dropwise via syringe over 10 min to a
precooled (-78 °C) solution of the cyclic sulfate shown in
Scheme 2 (2.00 g, 9.99 mmol) in THF (20 mL). The mixture was
stirred for 30 min at -78 °C and then allowed to slowly warm to
room temperature over 30 min. The mixture was cooled again to
-78 °C and treated with a solution of Ph2PLi {prepared from
Ph2PH (1.73 mL, 9.99 mmol) in THF (15 mL) and BuLi (7.40 mL
of a 1.6 M solution in hexane, 12.0 mmol) at -78 °C}. The mixture
was allowed to warm to room temperature and then stirred for
16 h. The solvent was then removed under reduced pressure, and
the residue was redissolved in Et2O (50 mL). Water (30 mL) was
added and the organic phase separated. The aqueous layer was
extracted with Et2O (3 ꢀ 20 mL). The combined organic phase
was dried over MgSO4, filtered, and concentrated to dryness
under reduced pressure. The residue was redissolved in the mini-
mum amount of pyrrolidine (15-20 mL) and then stirred for 16 h.
The pyrrolidine was removed under reduced pressure, and the oily
solidwas recrystallizedfrom the minimum amount (15-20 mL) of
boiling MeOH to give a white solid (1.65 g, 38%). Accurate mass
spectrum (ESI): Mr = 435.2368 (calcd for M þ H, C28H37P2
435.2365). 31P NMR (162 MHz, CD2Cl2): δ 26.3 (s), -14.9 (s). 1H
NMR (400 MHz, CD2Cl2): δ 7.53-7.34 (10 H, m), 7.12 (1 H, t,
J(HH) = 7.5 Hz), 6.95 (1 H, t, J(HH) = 7.5 Hz), 6.81 (1 H, d,
J(HH) = 7.5 Hz), 3.82 (2 H, s), 2.99 (2 H, s), 1.18 (18 H, d, J(HP) =
12.4 Hz). 13C NMR (100 MHz, CD2Cl2):δ139.9 (dd, J(CP) =8.6Hz,
J(CP) = 3.9 Hz), 139.1 (d, J(CP) = 15.6 Hz), 136.1 (dd, J(CP) = 6.9
Hz, J(CP) = 2.3 Hz), 133.6 (d, J(CP) = 18.7 Hz), 131.7 (dd, J(CP) =
11.7 Hz, J(CP) = 1.6 Hz), 131.3 (d, J(CP) = 7.8 Hz), 129.2, 128.9 (d,
J(CP) =6.2Hz),126.4(d,J(CP) =2.3Hz),125.8(t,J(CP) =1.9Hz),
34.3 (dd, J(CP) = 16.4 Hz, J(CP) = 9.3 Hz), 32.4 (d, J(CP) = 23.4
Hz), 30.3 (d, J(CP) = 13.2 Hz), 27.0 (dd, J(CP) = 25.7 Hz, J(CP) =
6.2 Hz).
Experimental Section
Synthesis of r-(Di-n-butylphosphino)-r0-(diphenylphosphino)-
xylene (L4). A 1.6 M solution of BuLi (2.81 mL, 4.55 mmol) in
hexane was added to a solution of Bu2PH BH3 (0.73 g, 4.5
Unless otherwise stated, all reactions were carried out under a
dry nitrogen atmosphere using standard Schlenk-line techni-
ques. Dry N2-saturated solvents were collected from a Grubbs
system38 in flame and vacuum-dried glassware. MeOH was
dried over 3 A molecular sieves, pentane was dried over 4 A
molecular sieves, and both were deoxygenated by N2 saturation.
The complexes [PtCl2(cod)],39 [Pt(CH3)2(cod)],40 [PtCl(CH3)-
(cod)],41 [PtH(Cl)(PPh3)2],42 [PdCl2(NCPh)2],43 [PdCl(CH3)-
(cod)],44 and [Pd(PCy3)2]21 were prepared by literature methods.
tBu2PH in THF was obtained from JCI USA Inc. and tBu2PH.
n
45
3
mmol) in THF (15 mL) at -78 °C. The mixture was allowed to
warm slowly to room temperature and recooled to -78 °C, and
then the nBu2P(BH3)Li solution was added dropwise via syringe
over 10 min to a precooled (-78 °C) solution of the cyclic sulfate
shown in Scheme 2 (0.76 g, 3.8 mmol) in THF (10 mL). The
mixture was stirred for 30 min at -78 °C and then stirred for
30 min at room temperature, cooled again to -78 °C, and
then treated with a solution of Ph2PLi {prepared from Ph2PH
(0.68 mL, 3.9 mmol) in THF (10 mL) and BuLi (2.40 mL of a
1.6 M solution in hexane, 3.88 mmol) at -78 °C}. The mixture
was allowed to warm to room temperature and then stirred for
(38) Pangborn, A. B.; Giardello, M. A.; Grubbs, R. H.; Rosen, R. K.;
Timmers, F. J. Organometallics 1996, 15, 1518.
(39) McDermott, J. X.; White, J.; Whitesides, G. M. J. Am. Chem.
Soc. 1976, 98, 6521.
(45) nBu2PH BH3 synthesized by a modification of the following
procedures: (a) Timmer, K.; Thewissen, D. H. M. W.; Marsman, J. W.;
3
(40) Costa, E.; Pringle, P. G.; Ravetz, M. Inorg. Synth. 1997, 31, 284.
(41) Clark, H. C.; Manzer, L. E. J. Organomet. Chem. 1973, 59, 411.
(42) Bailar, J. C.; Itatani, H. Inorg. Chem. 1965, 4, 1618.
(43) Anderson, G. K.; Lin, M. Inorg. Synth. 1990, 28, 60.
(44) Rulke, R. E.; Han, I. M; Elsevier, C. J.; Vrieze, K.; van Leeuwen,
P. W. M. N.; Roobeek, C. F.; Zoutberg, M. C.; Wang, Y. F.; Stam, C. H.
Inorg. Chim. Acta 1990, 169, 5.
Jan, W. Recl. Trav. Chim. Pays-Bas 1988, 107 (3), 248. (b) Naghipour, A.;
ꢀ
Sabounchei, S. J.; Morales-Morales, D.; Hernandez-Ortega, S.; Jensen, C. M.
J. Organomet. Chem. 2004, 689, 2494. 31P{1H} NMR (121 MHz, CDCl3): δ
-8.1 (q, J(BP) = 58 Hz). 1H NMR (300 MHz, CDCl3): δ 4.48 (HP, 1 H, br d,
J(HP) = 355.8 Hz), 1.77-1.29 (CH2, 12 H, m), 0.87 (CH3, 3 H, t, J(HH) =
7.2 Hz), 0.86 (CH3, 3 H, t, J(HH) = 7.2 Hz), 0.67-(-0.13) (BH3, 3 H, m).

Article
Organometallics, Vol. 29, No. 10, 2010 2303
16 h. The solvent was then removed under reduced pressure, and
the residue was redissolved in Et2O (20 mL). Water (20 mL) was
added and the organic layer separated. The aqueous phase was
extracted with Et2O (3 ꢀ 20 mL). The combined organic phase
was dried over MgSO4, filtered, and concentrated to dryness
under reduced pressure. The residue was redissolved in the
minimum amount of pyrrolidine (4 mL) and then heated for
16 h at 60 °C. The pyrrolidine was removed under reduced
pressure, and the oily solid was recrystallized from the minimum
amount (5 mL) of boiling MeOH to give a white solid (0.81 g,
49%). Accurate mass spectrum (ESI): Mr = 435.2376 (calcd for
M þ H, C28H37P2 435.2370). 31P NMR (121 MHz, C6D6): δ
(2 mL). The solution for stirred for 16 h, and then the solvent
was removed under reduced pressure. The solid was washed with
pentane to give a white solid (100 mg, 60%). Anal. Found (calcd
for C29H39ClP2Pt 0.5C5H12): C: 52.41 (52.10), H: 6.25 (6.20).
3
Accurate mass spectrum (ESI): Mr = 644.2165 (calcd for (M -
Cl)þ C29H39P2Pt 644.2169). For 9a: 31P NMR (121 MHz,
CD2Cl2): δ 31.3 (d, J(PPt) = 1828 Hz, J(PP) = 15 Hz), 7.4 (d,
J(PPt) = 4532 Hz, J(PP) = 15 Hz). 1H NMR (300 MHz,
CD2Cl2): δ 8.15-6.81 (13 H, m), 5.83 (1 H, d, J(HH) = 9.0
Hz), 4.55-3.34 (4 H, m), 1.82-1.30 (18 H, m), 0.25 (3 H, dd,
J(HPt) = 49.5 Hz, J(HP) = 6.6, 4.8 Hz). 13C NMR (75 MHz,
CD2Cl2): δ 136.1, 134.0, 133.9, 132.8, 131.8 (dd, J(CP) = 4.9 Hz,
J(CP) = 2.3 Hz), 130.8 (dd, J(CP) = 4.3 Hz, J(CP) = 3.2 Hz),
128.6, 128.1, 126.9, 126.0, 37.6 (J(CP) = 32.9 Hz), 37.0
(C(CH3)3), 31.7 (br s, CH3), 30.0 (br s, CH3), 26.9 (C(CH3)3),
26.1 (d, J(CP) = 12.7 Hz), 13.8 (CH3-Pt, dd, J(CP) = 87.7, 8.1
Hz). For 9b: 31P NMR (121 MHz, CD2Cl2): δ 31.4 (d, J(PPt) =
4267 Hz, J(PP) = 16 Hz), 4.3 (d, J(PPt) = 1704 Hz, J(PP) =
16 Hz). The 1H signals for 9b were obscured by the signals for 9a.
Synthesis of [PtH(PPh3)(L3)]Cl (12). A solution of [PtH(Cl)-
(PPh3)2] (21 mg, 0.03 mmol) in MeOH (1 mL) was added to a
suspension of L3 (12 mg, 0.03 mmol) in MeOH (1 mL). The dark
orange solution was stirred for 16 h, and the resulting mixture was
filtered to remove a precipitate. The solvent was then removed
under reduced pressure, giving a dark brown oil that contained a
mixture of products; the major one was characterized by NMR
spectroscopy. 31P NMR (121 MHz, CH3OH): δ 53.8 (PtBu2, dd,
J(PPt) = 2663 Hz, J(PP) = 315, 21 Hz), 21.3 (PPh3, dd, J(PPt) =
2610 Hz, J(PP) = 315, 21 Hz), 2.2 (PPh2, t, J(PPt) = 2226 Hz,
J(PP) = 21 Hz). 1H NMR (270 MHz, CD3OD): δ 7.64-6.93 (27
H, m), 6.74 (1 H, t, J(HH) = 7.7 Hz), 5.97 (1 H, d, J(HH) =
7.4 Hz), 4.28-4.08 (4 H, m), 1.41 (18 H, br s), -5.01 (HPt, ddd,
J(HPt) = 770 Hz, J(HP) = 158.0, 12.4, 11.4 Hz).
1
-14.1 (s), -28.9 (s). H NMR (300 MHz, C6D6): δ 7.44-6.75
(12 H, m), 3.78 (2 H, s), 2.91 (2 H, m), 1.42-1.20 (12 H, m),
0.86-0.75 (6 H, m). 13C NMR (75 MHz, C6D6): δ 139.6 (d,
J(CP) = 16.7 Hz), 138.1, 136.2 (d, J(CP) = 4.0 Hz), 133.8 (d,
J(CP) = 18.5 Hz), 131.7 (d, J(CP) = 6.9 Hz), 131.0 (d, J(CP) =
4.6 Hz), 129.1, 129.0, 128.9, 126.6 (d, J(CP) = 19.0 Hz), 34.2 (d,
J(CP) = 10.4 Hz), 34.0 (d, J(CP) = 8.7 Hz), 28.4 (d, J(CP) =
13.9 Hz), 27.9 (d, J(CP) = 15.0 Hz), 25.1 (d, J(CP) = 11.0 Hz),
14.4.
In Situ Generation of R2P(Se)CH2C6H4CH2P(Se)R02 (L1-4Se2).
Ligands L1-4 (25 mg) were added to a solution of KSeCN (100 mg)
in MeOH (1 mL) at room temperature. After 10 min, the MeOH
was evaporated and the crude product was dissolved in CDCl3. 31
P
NMR (121 MHz, CDCl3): L1(Se)2 δ 77.4 (J(PSe) = 694 Hz);
L2(Se)2 δ32.9 (J(PSe) =724Hz);L3(Se)2 δ75.7 (J(PSe) =694Hz,
J(PP) = 4 Hz), 33.0 (J(PSe) = 724 Hz, J(PP) = 4 Hz); L4(Se)2 δ
35.4 (J(PSe) = 687 Hz, J(PP) = 5 Hz), 33.6 (J(PSe) = 729 Hz,
J(PP) = 5 Hz).
Synthesis of [PtCl2(L3)] (3). A solution of L3 (100 mg, 0.23
mmol) in CH2Cl2 (2 mL) was added dropwise to a solution of
[PtCl2(cod)] (86 mg, 0.22 mmol) in CH2Cl2 (2 mL). The solutionwas
stirred for 16 h, and then the solvent was removed under reduced
pressure. The residue was washed with pentane and recrystallized
from CH2Cl2/hexane to give an off-white solid (91 mg, 56%). Anal.
Found (calcd for C28H36Cl2P2Pt): C: 48.14 (48.01), H: 5.28 (5.18).
ESI mass spectrum: m/z 664.2 (M - Cl)þ. 31P NMR (121 MHz,
CD2Cl2): δ 25.0 (d, J(PPt) = 3603 Hz, J(PP) = 13 Hz), -7.0 (d,
J(PPt) = 3606 Hz, J(PP) = 13 Hz). 1H NMR (400 MHz, C2D2Cl4):
δ 8.08-6.79 (13 H, m), 5.92 (1 H, d, J(HH) = 7.8 Hz), 3.96-3.79
(1 H, m), 3.60-3.41 (1 H, m), 2.66-2.59 (1 H, m), 2.23-2.17 (1 H,
m), 1.68-0.83 (18 H, m). 13C NMR (75 MHz, CD2Cl2): δ 134.7,
134.4, 132.4, 132.1, 131.6, 131.1, 129.1, 127.9, 127.3 100.7, 36.9 (d,
J(CP) = 31.2 Hz), 32.7 (br s, C(CH3)3), 31.4 (C(CH3)3), 30.5 (br s,
C(CH3)3), 28.5 (C(CH3)3), 26.1 (d, J(CP) = 26.5 Hz).
Synthesis of [PdCl2(L3)] (13). A solution of L3 (50 mg, 0.12
mmol) in CH2Cl2 (2 mL) was added dropwise over 30 s to a
stirred solution of [PdCl2(NCPh)2] (44 mg, 0.12 mmol) in
CH2Cl2 (2 mL). The solution was stirred overnight, and then
the volatiles were removed under reduced pressure to give the
crude yellow solid product, which was crystallized from
CH2Cl2/hexane (62 mg, 58%). Anal. Found (calcd for
C28H36Cl2P2Pd 0.75CH2Cl2): C: 51.56 (51.10), H: 5.62 (5.55).
3
Accurate mass spectrum (ESI): Mr = 575.1025 (M - Cl)þ (calcd
for C28H36ClP2Pd 575.1010). 31P NMR (121 MHz, CD2Cl2): δ
47.8 (d, J(PP) = 4 Hz), 13.1 (d, J(PP) = 4 Hz). 1H NMR (400
MHz, CD2Cl2): δ 8.20-7.26 (11 H, m), 7.16 (1 H, t, J(HH) =
7.7 Hz), 6.81 (1 H, t, J(HH) = 7.4 Hz), 5.94 (1 H, d, J(HH) =
7.0 Hz), 4.04-3.18 (CH2, 4 H, m), 1.87-1.50 (CH3, 18 H, m).
13C NMR (100 MHz, CD2Cl2): δ 134.2 (br s), 133.3 (dd,
J(CP) = 4.7 Hz, J(CP) = 3.9 Hz), 131.7-131.6 (2 C, m), 131.0
(d, J(CP) = 8.6 Hz), 130.7 (dd, J(CP) = 4.3, J(CP) = 3.1 Hz),
128.4 (d, J(CP) = 11.3 Hz), 127.7 (br s), 127.5 (dd, J(CP) =3.5Hz,
J(CP) = 2.1 Hz), 126.9 (t, J(CP) = 2.8 Hz), 40.8 (C(CH3)3, br s),
39.4 (C(CH3)3, br s), 36.8 (CH2, dd, J(CP) = 22.6 Hz, J(CP) = 3.1
Hz), 31.9 (C(CH3)3, br s), 30.3 (C(CH3)3, br s), 26.0 (CH2, dd,
J(CP) = 15.6 Hz, J(CP) = 3.1 Hz).
Synthesis of [Pt(CH3)2(L3)] (6). A solution of L3 (33 mg, 0.08
mmol) in CH2Cl2 (0.5 mL) was added dropwise to a solution of
[Pt(CH3)2(cod)] (25 mg, 0.08 mmol) in CH2Cl2 (0.5 mL). The
solution was stirred for 16 h, and then the solvent was removed
under reduced pressure. The residue was recrystallized from
CH2Cl2/hexane to give a white solid (36 mg, 69%). Accurate
mass spectrum (ESI): Mr = 644.2200 (calcd for (M - CH3)þ
C29H39P2Pt 644.2169). 31P NMR (121 MHz, CD2Cl2): δ 27.6 (d,
J(PPt) = 1883 Hz, J(PP) = 15 Hz), 5.4 (d, J(PPt) = 1929 Hz,
1
J(PP) = 15 Hz). H NMR (400 MHz, CD2Cl2): δ 8.02-7.05
Synthesis of [PdCl(CH3)(L3)] (14a/14b). A solution of L3
(50 mg, 0.12 mmol) in CH2Cl2 (2 mL) was added dropwise to
a solution of [PdCl(CH3)(cod)] (31 mg, 0.12 mmol) in CH2Cl2
(2 mL). The solution was stirred for 16 h, and then the solvent
was removed under reduced pressure. The residue was recrys-
tallized from CH2Cl2/hexane to give a yellow solid (30 mg,
42%). Accurate mass spectrum (ESI): Mr = 555.1579 (calcd for
(M - Cl)þ C29H39P2Pd 555.1556). For 14a: 31P NMR (121 MHz,
CD2Cl2): δ 28.9 (d, J(PP) = 37 Hz), 28.4 (d, J(PP) = 37 Hz). 1H
NMR (300 MHz, CD2Cl2): δ 8.06-7.07 (12 H, m), 6.80 (1 H, t,
J(HH) = 7.2 Hz), 5.99 (1 H, d, J(HH) = 7.7 Hz), 4.06-3.36 (4 H,
m), 1.5 (18 H, br s), 0.54 (3 H, dd, J(HP) = 5.9, 3.5 Hz). 13C NMR
(75 MHz, CD2Cl2): δ 136.4, 134.0 (d, J(CP) = 10 Hz), 132.6, 132.0,
131.4, 131.1, 130.8, 128.9 (br s), 126.9, 126.0, 38.0 (dd, J(CP) =
21 Hz, J(CP) = 6.6 Hz), 36.9 (C(CH3)3), 36.3 (C(CH3)3), 31.1
(12 H, m), 6.77 (1 H, t, J(HH) = 7.5 Hz), 6.02 (1 H, d, J(HH) =
7.5 Hz), 4.12-3.45 (4 H, br m), 1.45 (18 H, br s), 0.69 (3 H, t,
J(HPt) = 69.9 Hz, J(HP) = 6.5 Hz), -0.03 (3 H, dd, J(HPt) =
65.3 Hz, J(HP) = 8.8, 6.3 Hz). 13C NMR (100 MHz, CD2Cl2): δ
137.0 (d, J(CP) = 3.8 Hz), 134.5 (d, J(CP) = 2.3 Hz), 131.6 (dd,
J(CP) = 4.6 Hz, J(CP) = 2.3 Hz), 131.0 (d, J(CP) = 3.1 Hz),
130.9, 129.6, 129.1, 128.3, 126.1 (dd, J(CP) = 3.1 Hz, J(CP) =
2.3 Hz), 125.6 (t, J(CP) = 2.3 Hz), 38.2 (CH2, d, J(CP) = 16.9
Hz), 37.5 (C(CH3)3), 37.3 (C(CH3)3), 31.6 (CH3), 30.0 (CH3),
27.7 (d, J(CP) = 11.5 Hz), 9.2 (CH3-Pt, dd, J(CP) = 94.5, 8.1
Hz), 3.7 (CH3-Pt, dd, J(CP) = 103.0, 7.7 Hz).
Synthesis of [PtCl(CH3)(L3)] (9a/9b). A solution of L3
(107 mg, 0.25 mmol) in CH2Cl2 (3 mL) was added dropwise to
a solution of [PtCl(CH3)(cod)] (87 mg, 0.25 mmol) in CH2Cl2

2304 Organometallics, Vol. 29, No. 10, 2010
Fanjul et al.
(CH3), 30.4 (CH3), 26.2 (d, J(CP) = 4.3 Hz), 19.2 (CH3-Pd, dd,
J(CP) = 80.0, 10.4 Hz). For 14b: 31P NMR (121 MHz, CD2Cl2): δ
29.5 (d, J(PP) = 36 Hz), 28.6 (d, J(PP) = 36 Hz). 1H NMR signals
for 14b obscured by signals for 14a.
Preparation of [PdH(PCy3)(L3)]BF4 (15). A solution of L3
(32 mg, 0.05 mmol) in THF (0.6 mL) was added to a solution of
[Pd(PCy)3] (21 mg, 0.05 mmol) in THF (0.4 mL). The orange
CD2Cl2 was added to dissolve the complex and the mixture was
stirred for 1 h. 31P NMR (121 MHz, CD2Cl2): δ 41.3 (dd, J(PPt) =
1749 Hz, J(PP) = 23 Hz, J(13CP) = 8 Hz), 6.1 (dd, J(PPt) = 3312
Hz, J(13CP) = 137, 23 Hz). 13C NMR (100 MHz, CD2Cl2): δ180.6
(Pt(13CO), dd, J(13CPt) = 1365.8 Hz, J(13CP) = 136.8, 8.9 Hz),
134.9 (dd, J(CP) = 8.5 Hz, J(CP) = 3.0 Hz), 133.8 (d, J(CP) =
18.4 Hz), 132.8 (d, J(CP) = 2.8 Hz), 132.1, 132.0 (d, J(CP) = 3.2
Hz), 131.0 (d, J(CP) = 1.8 Hz), 128.7, 128.2 (dd, J(CP) = 8.8 Hz,
J(CP) = 1.4 Hz), 127.7 (dd, J(CP) = 15.7 Hz, J(CP) = 2.8 Hz),
127.1 (d, J(CP) = 2.8 Hz), broad signals observed in the region
40.8-25.1, 5.9 (br s). ΙR (cm-1): νCO 2052.
Preparation of [Pt(CH3)(NCMe)(L3)]BF4 (22a). Mixture
9a/9b (64 mg, 0.087 mmol) was dissolved in a 1:1 mixture of CH2Cl2
and MeCN (4 mL). AgBF4 (19 mg, 0.096 mmol) was added to the
solution, and the reaction mixture was stirred for 16 h. The solution
was filtered through a Celite pad and the Celite washed with CH2Cl2.
The solvent was then removed under reduced pressure to give a
white solid. Yield: 47 mg, 67%. 31P NMR (121 MHz, CD2Cl2): δ
35.6 (d, J(PPt) = 1824 Hz, J(PP) = 17 Hz), 4.1 (d, J(PPt) = 4598
Hz, J(PP) = 17 Hz). 1H NMR (300 MHz, CD2Cl2): δ 8.04-7.15
(12 H, m), 6.88 (1 H, t, J(HH) = 7.6 Hz), 6.06 (1 H, d, J(HH) =
7.4 Hz), 4.05-3.48 (4 H, br m), 2.51 (3 H, d, J(HPt) = 8.4 Hz), 1.44
(18 H, d, J(HP) = 13.4 Hz), 0.29 (3 H, dd, J(HPt) = 47 Hz,
J(HP) = 6.1, 4.0 Hz). 13C NMR (68 MHz, CD2Cl2): δ 134.4 (dd,
J(CP) = 4.3 Hz, J(CP) = 1.8 Hz), 134.0, 133.8 (d, J(CP) = 10.4
Hz), 132.1 (dd, J(CP) = 5.2 Hz, J(CP) =2.6 Hz), 131.8 (br s), 131.4
(dd, J(CP) = 4.9 Hz, J(CP) = 2.8 Hz), 131.1 (dd, J(CP) = 4.7 Hz,
J(CP) = 3.1 Hz), 128.6 (br s), 127.7 (dd, J(CP) = 3.6 Hz, J(CP) =
2.1 Hz), 126.9 (t, J(CP) = 3.1 Hz), 37.7 (C(CH3)3), 37.6 (C(CH3)3),
38.5 (dd, J(CP) = 36.3 Hz, J(CP) = 3.1 Hz), 30.4 (CH3), 29.7
(CH3), 25.3 (d, J(CP) = 17.7 Hz), 9.3 (CH3-Pt, dd, J(CP) = 74, 6.8
Hz), 4.0, 1.7.
solution was stirred for 30 min, and HBF4 OEt2 was added
3
(7 μL, 0.05 mmol) to the reaction mixture. The red solution was
then stirred for 1 h. After removal of the volatiles in vacuo, the
product was recrystallized from THF/hexane to give a white
solid. Accurate mass spectrum (ESI): Mr = 821.3722 (calcd for
Mþ C46H70P3Pd 821.3720). 31P NMR (121 MHz, CD2Cl2): δ
58.7 (PtBu2, dd, J(PP) = 302, 34 Hz), 38.2 (PCy3, dd, J(PP) =
302, 26 Hz), 1.1 (PPh2, dd, J(PP) = 34, 26 Hz). 1H NMR (270
MHz, CD2Cl2): δ 7.88-7.36 (11 H, m), 7.11 (1 H, t, J(HH) =
7.6 Hz), 6.77 (1 H, t, J(HH) = 7.4 Hz), 5.97 (1 H, d, J(HH) =
7.4 Hz), 3.86-3.63 (4 H, m), 1.94-0.83 (51 H, m), -7.85 (H-Pd,
1 H, br dd, J(HP) = 182.7, 13.9 Hz).
Synthesis of [PtCl2(L4)] (16). A solution of L4 (100 mg, 0.23
mmol) in CH2Cl2 (1 mL) was added dropwise to a solution of
[PtCl2(cod)] (78 mg, 0.21 mmol) in CH2Cl2 (2 mL). The solution
was stirred for 16 h, and then the solvent was removed under
reduced pressure. The residue was washed with pentane to give a
white solid (92 mg, 63%). Anal. Found (calcd for C28H36Cl2-
P2Pt 0.25C5H12): C: 48.85 (48.84), H: 5.44 (5.43). Accurate
3
mass spectrum (ESI): Mr = 664.1649 (calcd for (M - Cl)þ,
C28H36ClP2Pt 664.1634). 31P NMR (121 MHz, CD2Cl2): δ -1.7
(d, J(PPt) = 3397 Hz, J(PP) = 12 Hz), -3.3 (d, J(PPt) = 3655
Hz, J(PP) = 12 Hz). 1H NMR (400 MHz, CD2Cl2): δ 7.85-7.81
(2 H, m), 7.54-7.44 (4 H, m), 7.17 (2 H, d, J(HH) = 4.4 Hz), 6.89
(1 H, m), 6.13 (1 H, dd, J(HH) = 7.4 Hz, J(HH) = 1.5 Hz),
3.94-3.15 (4 H, m), 2.46-0.86 (18 H, m). 13C NMR (100 MHz,
CD2Cl2): δ 134.7 (d, J(CP) = 9.3 Hz), 133.0 (d, J(CP) = 3.9 Hz),
131.8 (d, J(CP) = 2.3 Hz), 131.6 (d, J(CP) = 3.1 Hz), 131.3 (dd,
J(CP) = 5.5 Hz, J(CP) = 3.9 Hz), 130.7 (dd, J(CP) = 5.5 Hz,
J(CP) = 3.1 Hz), 128.7, 128.6, 128.3 (t, J(CP) = 3.1 Hz), 127.8
(t, J(CP) = 3.5 Hz), 36.7 (d, J(CP) = 35.0 Hz), 31.0 (d, J(CP) =
31.1 Hz), 28.2 (d, J(CP) = 2.3 Hz), 27.3 (d, J(CP) = 39.7 Hz),
24.5 (d, J(CP) = 15.6 Hz), 14.0.
Reaction of [PdCl(CH3)(L3)] (14a/14b) with 13CO. The mix-
ture 14a/14b (100 mg, 0.17 mmol) was placed in a three-necked
flask connected to a 13CO gas cylinder via a Schlenk line. When
the system was under 13CO atmosphere, CD2Cl2 was added to
dissolve the complex and the mixture was stirred for 1.5 h, and
then the products 23a and 23b were identified in solution. For
23a: 31P NMR (121 MHz, CD2Cl2): δ 27.8 (dd, J(13CP) = 103,
1
61 Hz), 15.5 (br d, J(PP) = 61 Hz, J(13CP) not resolved). H
Synthesis of [PtCl(CH3)(L4)] (17a/17b). A solution of L4
(126 mg, 0.29 mmol) in CH2Cl2 (1.4 mL) was added dropwise
to a solution of [PtCl(CH3)(cod)] (96 mg, 0.27 mmol) in CH2Cl2
(3 mL). The solution was stirred for 16 h, and then the solvent
was removed under reduced pressure. The solid was washed with
pentane to give a light brown solid (100 mg, 55%). Accurate
mass spectrum (ESI): Mr = 664.2197 (calcd for (M - Cl)þ,
C29H39P2Pt 664.2169). For 17a: 31P NMR (121 MHz, CD2Cl2):
δ 7.9 (d, J(PPt) = 4412 Hz, J(PP) = 16 Hz), 3.1 (d, J(PPt) =
NMR (300 MHz, CD2Cl2): δ 7.75-6.73 (14 H, m), 4.46-2.35 (4
H, m), 1.79 (3 H, dd, J(H13C) = 5.7 Hz, J(HP) = 0.9 Hz), 1.54 (9
H, d, J(HP) = 12.7 Hz), 1.19 (9 H, d, J(HP) = 12.7 Hz).
13C NMR (100 MHz, CD2Cl2): δ 232.1 (Pd-13C(O)CH3, dd,
J(13CP) = 103.0, 13.8 Hz), 136.3 (dd, J(CP) = 3.7 Hz, J(CP) =
1.8 Hz), 132.2, 132.1, 131.7 (d, J(CP) =2.8Hz),131.3(J(CP) =3.7
Hz), 131.2 (J(CP) = 9.2 Hz), 128.7, 128.5 (J(CP) = 12.0 Hz), 127.0
(dd, J(CP) = 3.7 Hz, J(CP) = 1.8 Hz), 126.2 (t, J(CP) = 2.7 Hz),
37.5 (Pd-13C(O)(CH3), ddd, J(13CC) = 22.1 Hz, J(CP) = 3.7, 1.4
Hz), 36.1 (C(CH3)3), 35.6 (d, J(CP) = 4.6 Hz), 35.2 (dd, J(CP) =
32.3 Hz, J(CP) = 5.5 Hz), 30.5 (C(CH3)3), 28.1 (C(CH3)3),
27.0 (C(CH3)3). ΙR (cm-1): νCO 1655 (br). For 23b: 13C NMR
(100 MHz, CD2Cl2):δ232.0 (Pd-13C(O)CH3, dd, J(13CP) = 103.3,
12.9 Hz). 31P NMR signals were not observed for the minor
product 23b above the noise level.
Carbonylation Catalysis. Using standard Schlenk line techni-
ques, reaction solutions were prepared by dissolving Pd(OAc)2
(22 mg, 0.10 mmol) and ligand L1-3 (0.50 mmol) in methanol
(300 mL). The mixture was stirred for 30 min after the addition
of of methane sulfonic acid (2.92 mL, 45 mmol) to complete the
preparation of the catalyst solution. The catalytic solution was
added to the pre-evacuated autoclave and heated to 100 °C, at
which point the pressure generated by the solvent was 2.3 bar.
The autoclave was then pressured to 12.3 bar with CO/ethene
(1:1) from a 10 L reservoir at higher pressure. A regulatory valve
ensured that the pressure of the autoclave was maintained
throughout the reaction at 12.3 bar through continual injection
of gas from the reservoir. The pressure of the reservoir as well as
the reactor temperature was logged throughout the reaction
period of 3 h. The amount of product at any point in the reaction
1
1766 Hz, J(PP) = 16 Hz). H NMR (300 MHz, CD2Cl2): δ
7.78-6.74 (14 H, m), 6.17 (1 H, d, J(HH) = 7.2 Hz), 4.18-3.13
(4 H, m), 2.36-0.76 (18 H, m), 0.11 (3 H, dd, J(HPt) = 51.0 Hz,
J(HP) = 7.0, 4.4 Hz). For 17b: 31P NMR (121 MHz, CD2Cl2): δ
5.8 (d, J(PPt) = 1779 Hz, J(PP) = 13 Hz), 1.0 (d, J(PPt) = 4059
Hz, J(PP) = 13 Hz). 1H NMR (300 MHz, CD2Cl2): δ 7.78-6.74
(14 H, m), 5.56 (1 H, s), 4.18-3.13 (4 H, m), 2.36-0.76 (18 H, m),
0.53 (3 H, dd, J(HPt) = 60.9 Hz, J(HP) = 7.3, 4.2 Hz). 13C
NMR (75 MHz, CD2Cl2): δ 134.7, 134.5, 133.8, 133.5, 133.2,
132.8, 132.7, 132.0, 131.9, 130.9, 130.8, 130.3, 130.3, 130.1,
128.6, 128.2 (d, J(CP) = 11.0 Hz), 127.6 (d, J(CP) = 9.2 Hz),
127.2, 126.8, 126.5, 38.1 (d, J(CP) = 37.8 Hz), 31.2 (d, J(CP) =
35.8 Hz), 28.5, 28.4, 27.9 (d, J(CP) = 30.0 Hz), 27.2 (d, J(CP) =
36.3 Hz), 26.5 (d, J(CP) = 24.2 Hz), 25.7 (d, J(CP) = 28.3 Hz),
24.9 (d, J(CP) = 12.7 Hz), 24.5 (d, J(CP) = 14.4 Hz), 14.1, 14.0,
8.9 (CH3-Pt, dd, J(CP) = 92.3, 7.5 Hz), 5.6 (CH3-Pt, dd, J(CP) =
96.9, 8.1 Hz).
Carbonylation Experiments. Synthesis of [Pt(CH3)(13CO)-
(L3)]BF4 (20a0). Complex 22a (47 mg, 0.060 mmol) was placed in
a three-necked flask connected to a 13CO gas cylinder via a
Schlenk line. When the system was under the 13CO atmosphere,

Article
Organometallics, Vol. 29, No. 10, 2010 2305
could be calculated from the drop in reservoir pressure by
assuming ideal gas behavior and 100% selectivity for methyl
propionate, allowing reaction TON with the particular ligand to
be estimated. After 3 h, the autoclave was cooled and vented.
The reaction solution was collected from the base of the vessel
and immediately placed under a N2 atmosphere.46 The solution
was weighed so that the TON could be calculated. In examples
where recycling was undertaken, the solution was then reduced
to 20 mL by evaporation under reduced pressure. This concen-
trated solution was left to stand for 16 h under an inert atmo-
sphere and was then, following addition of methanol (300 mL),
used as the catalyst solution under the same set of conditions as
the original. The catalyst was recycled in this way, until a
significant drop in reaction TON was observed.
Crystal Structure Determinations. Data were collected using
Mo KR X-radiation (λ = 0.71073 A) on single crystals mounted
on a glass fiber with paraffin oil. Data for crystal samples,
[PtCl2(L3)] (3), [PdCl(CH3)(L3)] (14a), and [PtCl2(L4)] (16) were
collected on a Bruker Apex II CCD detector diffractometer.
Data for [PdH(CH3)(PCy3)(L3)][BF4] (15) were collected at the
National Crystallography Service (NCS), Southampton, with a
Bruker-Nonius APEX II CCD camera on a κ-goniostat dif-
fractometer with a rotating anode source using 10 cm confocal
mirrors. Structure solution and refinement were performed
using SHELXTL software. All atoms were refined with aniso-
tropic displacement parameters expect the hydrogen atoms,
which were isotropic and were refined in constrained geometries
with fixed isotropic displacement parameters.
Computational Details. All calculations described here were
performed with the standard PW91 density functional47 as
implemented in Jaguar 6.048 with a standard 6-31G* basis set
on all atoms apart from the metal centers, where the Jaguar
triple-ζ form of the standard Los Alamos ECP basis set
(LACV3P) was used. Optimizations were performed with a
self-consistent reaction field dielectric continuum49 for metha-
nol (ε = 33.62, radprb = 2.002) as implemented in Jaguar.
Calculations used “loose” geometry convergence criteria
(5 times larger than default criteria) to accelerate structural
convergence; test calculations with the more stringent default
criteria did not lead to substantial changes in energies or
structures, but were much more time-consuming. Tighter DFT
grids (gdftgrad, gdftmed, and gdftfine = -13) with tight cutoffs
(iacc = 1) were requested for the SCF calculations. The effect of
different DFT functionals, basis sets, and optimization in vacuo
on structures and isomer energies has also been explored, and
details of this can be found in the Supporting Information.
Acknowledgment. We are grateful to Lucite and
EPSRC for studentships and an Advanced Research
Fellowship to NF (EP/E059376/F), the Royal Society,
COST action CM0802 “PhoSciNet”, for supporting this
work, and Johnson Matthey for a loan of precious metal
compounds.
(46) Exposure of the solutions to air is kept to a minimum during
these recycle procedures, but oxygen was not rigorously excluded.
However, the reaction mixtures contain 90 equiv of methane sulfonic
acid and under these conditions the P atoms (particularly of the
highly basic tBu2P groups) are protonated and thereby protected from
oxidation.
(47) (a) Slater, J. C. Quantum Theory of Molecules and Solids, Vol. 4:
The Self-Consistent Field for Molecules and Solids; McGraw-Hill: New
York, 1974. (b) Perdew, J. P. In Electronic Structure Theory of Solids;
Ziesche, P., Eschrig, H., Eds.;Akademie Verlag: Berlin, 1991. (c) Perdew, J.
P.; Chevary, J. A.; Vosko, S. H.; Jackson, K. A.; Pederson, M. R.; Singh, D. J.;
Fiolhais, C. Phys. Rev. B 1992, 46, 6671.
Supporting Information Available: Cif files reporting crystal
and refinement data for all crystal structures. Details of the
effect of different DFT functionals, basis sets, and optimization
in vacuo on structures and isomer energies. This material is
available free of charge via the Internet at http://pubs.acs.org.
(49) (a) Tannor, D. J.; Marten, B.; Murphy, R.; Friesner, R. A.;
Sitkoff, D.; Nicholls, A.; Ringnalda, M.; Goddard, W. A., III; Honig, B.
J. Am. Chem. Soc. 1994, 116, 11875. (b) Marten, B.; Kim, K.; Cortis, C.;
Friesner, R. A.; Murphy, R. B.; Ringnalda, M. N.; Sitkoff, D.; Honig, B. J.
Phys. Chem. 1996, 100, 11775.
€
(48) Jaguar, version 6.0; Schrodinger, L.L.C.: New York, 2005.